├── maskrcnn_benchmark ├── utils │ ├── __init__.py │ ├── README.md │ ├── miscellaneous.py │ ├── collect_env.py │ ├── cv2_util.py │ ├── logger.py │ ├── imports.py │ ├── timer.py │ ├── env.py │ └── registry.py ├── modeling │ ├── __init__.py │ ├── roi_heads │ │ ├── __init__.py │ │ ├── box_head │ │ │ └── __init__.py │ │ ├── keypoint_head │ │ │ ├── __init__.py │ │ │ └── roi_keypoint_predictors.py │ │ └── mask_head │ │ │ └── __init__.py │ ├── rpn │ │ ├── retinanet │ │ │ └── __init__.py │ │ ├── __init__.py │ │ └── utils.py │ ├── detector │ │ ├── __init__.py │ │ └── detectors.py │ ├── backbone │ │ └── __init__.py │ ├── registry.py │ └── utils.py ├── structures │ └── __init__.py ├── __init__.py ├── engine │ └── __init__.py ├── config │ └── __init__.py ├── data │ ├── __init__.py │ ├── datasets │ │ ├── __init__.py │ │ ├── evaluation │ │ │ ├── voc │ │ │ │ └── __init__.py │ │ │ ├── coco │ │ │ │ └── __init__.py │ │ │ └── __init__.py │ │ ├── concat_dataset.py │ │ └── list_dataset.py │ ├── transforms │ │ ├── __init__.py │ │ └── build.py │ ├── samplers │ │ ├── __init__.py │ │ └── iteration_based_batch_sampler.py │ └── collate_batch.py ├── solver │ ├── __init__.py │ └── build.py ├── layers │ ├── nms.py │ ├── smooth_l1_loss.py │ ├── __init__.py │ ├── batch_norm.py │ └── _utils.py └── csrc │ ├── cpu │ └── vision.h │ ├── vision.cpp │ ├── nms.h │ ├── SigmoidFocalLoss.h │ ├── ROIPool.h │ └── ROIAlign.h ├── matlab └── DenseCRF │ ├── .gitignore │ ├── include │ ├── densecrf │ │ ├── external │ │ │ ├── liblbfgs │ │ │ │ ├── NEWS │ │ │ │ ├── AUTHORS │ │ │ │ ├── Makefile.am │ │ │ │ ├── sample │ │ │ │ │ └── Makefile.am │ │ │ │ ├── lib │ │ │ │ │ └── Makefile.am │ │ │ │ ├── autogen.sh │ │ │ │ ├── COPYING │ │ │ │ ├── lbfgs.sln │ │ │ │ └── config.h.in │ │ │ └── CMakeLists.txt │ │ ├── include │ │ │ └── Eigen │ │ │ │ ├── src │ │ │ │ ├── SparseCore │ │ │ │ │ ├── SparseAssign.h │ │ │ │ │ ├── CMakeLists.txt │ │ │ │ │ ├── SparseFuzzy.h │ │ │ │ │ └── SparseRedux.h │ │ │ │ ├── Core │ │ │ │ │ ├── util │ │ │ │ │ │ ├── NonMPL2.h │ │ │ │ │ │ ├── CMakeLists.txt │ │ │ │ │ │ └── ReenableStupidWarnings.h │ │ │ │ │ ├── arch │ │ │ │ │ │ ├── CMakeLists.txt │ │ │ │ │ │ ├── NEON │ │ │ │ │ │ │ └── CMakeLists.txt │ │ │ │ │ │ ├── SSE │ │ │ │ │ │ │ └── CMakeLists.txt │ │ │ │ │ │ ├── AltiVec │ │ │ │ │ │ │ └── CMakeLists.txt │ │ │ │ │ │ └── Default │ │ │ │ │ │ │ └── CMakeLists.txt │ │ │ │ │ ├── products │ │ │ │ │ │ └── CMakeLists.txt │ │ │ │ │ └── CMakeLists.txt │ │ │ │ ├── QR │ │ │ │ │ └── CMakeLists.txt │ │ │ │ ├── SVD │ │ │ │ │ └── CMakeLists.txt │ │ │ │ ├── misc │ │ │ │ │ └── CMakeLists.txt │ │ │ │ ├── Jacobi │ │ │ │ │ └── CMakeLists.txt │ │ │ │ ├── LU │ │ │ │ │ ├── arch │ │ │ │ │ │ └── CMakeLists.txt │ │ │ │ │ └── CMakeLists.txt │ │ │ │ ├── plugins │ │ │ │ │ └── CMakeLists.txt │ │ │ │ ├── Cholesky │ │ │ │ │ └── CMakeLists.txt │ │ │ │ ├── Eigenvalues │ │ │ │ │ └── CMakeLists.txt │ │ │ │ ├── Householder │ │ │ │ │ └── CMakeLists.txt │ │ │ │ ├── StlSupport │ │ │ │ │ └── CMakeLists.txt │ │ │ │ ├── Geometry │ │ │ │ │ ├── arch │ │ │ │ │ │ └── CMakeLists.txt │ │ │ │ │ └── CMakeLists.txt │ │ │ │ ├── PaStiXSupport │ │ │ │ │ └── CMakeLists.txt │ │ │ │ ├── CholmodSupport │ │ │ │ │ └── CMakeLists.txt │ │ │ │ ├── PardisoSupport │ │ │ │ │ └── CMakeLists.txt │ │ │ │ ├── SparseCholesky │ │ │ │ │ └── CMakeLists.txt │ │ │ │ ├── SuperLUSupport │ │ │ │ │ └── CMakeLists.txt │ │ │ │ ├── UmfPackSupport │ │ │ │ │ └── CMakeLists.txt │ │ │ │ ├── OrderingMethods │ │ │ │ │ └── CMakeLists.txt │ │ │ │ ├── Eigen2Support │ │ │ │ │ ├── Geometry │ │ │ │ │ │ └── CMakeLists.txt │ │ │ │ │ ├── CMakeLists.txt │ │ │ │ │ ├── Macros.h │ │ │ │ │ ├── TriangularSolver.h │ │ │ │ │ └── Memory.h │ │ │ │ ├── IterativeLinearSolvers │ │ │ │ │ └── CMakeLists.txt │ │ │ │ └── CMakeLists.txt │ │ │ │ ├── Eigen │ │ │ │ ├── Dense │ │ │ │ ├── Array │ │ │ │ ├── Sparse │ │ │ │ ├── OrderingMethods │ │ │ │ ├── Householder │ │ │ │ ├── CMakeLists.txt │ │ │ │ ├── Jacobi │ │ │ │ ├── StdList │ │ │ │ ├── QtAlignedMalloc │ │ │ │ ├── LeastSquares │ │ │ │ ├── StdDeque │ │ │ │ ├── StdVector │ │ │ │ ├── Cholesky │ │ │ │ ├── PardisoSupport │ │ │ │ ├── SparseCholesky │ │ │ │ ├── SVD │ │ │ │ ├── QR │ │ │ │ ├── LU │ │ │ │ ├── UmfPackSupport │ │ │ │ ├── Eigenvalues │ │ │ │ ├── PaStiXSupport │ │ │ │ ├── IterativeLinearSolvers │ │ │ │ ├── Geometry │ │ │ │ └── CholmodSupport │ │ ├── examples │ │ │ ├── im1.ppm │ │ │ ├── im2.ppm │ │ │ ├── im3.ppm │ │ │ ├── anno1.ppm │ │ │ ├── anno2.ppm │ │ │ ├── anno3.ppm │ │ │ ├── CMakeLists.txt │ │ │ ├── common.h │ │ │ └── ppm.h │ │ ├── CMakeLists.txt │ │ └── src │ │ │ └── CMakeLists.txt │ ├── cppmatrix.cpp │ ├── energy_mex.mexa64 │ ├── densecrf_mex.mexa64 │ ├── TRW_S-v1.3 │ │ ├── LICENSE.TXT │ │ ├── CHANGES.TXT │ │ ├── instances.inc │ │ ├── instances.h │ │ ├── README.TXT │ │ └── treeProbabilities.cpp │ ├── probImage │ │ ├── compress_zip.sh │ │ ├── decompress_zip.sh │ │ ├── compress_all.sh │ │ ├── CMakeLists.txt │ │ ├── compress.cpp │ │ ├── decompress.cpp │ │ ├── decompress_all.sh │ │ └── probimage.h │ ├── solvers.h │ ├── maxflow-v3.03.src │ │ ├── instances.inc │ │ └── CHANGES.TXT │ ├── decompress.m │ ├── decompress_mex.cpp │ ├── extendedDenseCRF2D.h │ └── compile.m │ ├── examples │ ├── img_1001.jpg │ ├── img_1001_stage1.png │ ├── img_1001_stage2.png │ ├── example_MSRC.m │ ├── example.m │ ├── SalRefineTest.m │ └── MyRun.m │ ├── screenshot │ └── screenshot.png │ ├── data │ └── readme.txt │ ├── LICENSE.md │ └── README.md ├── requirements.txt ├── fig └── intro.png ├── demo ├── demo_e2e_mask_rcnn_R_50_FPN_1x.png ├── demo_e2e_mask_rcnn_X_101_32x8d_FPN_1x.png └── README.md ├── configs ├── caffe2 │ ├── e2e_faster_rcnn_R_50_C4_1x_caffe2.yaml │ ├── e2e_mask_rcnn_R_50_C4_1x_caffe2.yaml │ ├── e2e_faster_rcnn_R_101_FPN_1x_caffe2.yaml │ ├── e2e_faster_rcnn_R_50_FPN_1x_caffe2.yaml │ ├── e2e_faster_rcnn_X_101_32x8d_FPN_1x_caffe2.yaml │ ├── e2e_mask_rcnn_R_50_FPN_1x_caffe2.yaml │ ├── e2e_mask_rcnn_R_101_FPN_1x_caffe2.yaml │ ├── e2e_mask_rcnn_X_101_32x8d_FPN_1x_caffe2.yaml │ ├── e2e_mask_rcnn_X-152-32x8d-FPN-IN5k_1.44x_caffe2.yaml │ └── e2e_keypoint_rcnn_R_50_FPN_1x_caffe2.yaml ├── rpn_R_50_C4_1x.yaml ├── e2e_faster_rcnn_R_50_C4_1x.yaml ├── pascal_voc │ ├── e2e_faster_rcnn_R_50_C4_1x_1_gpu_voc.yaml │ ├── e2e_faster_rcnn_R_50_C4_1x_4_gpu_voc.yaml │ └── e2e_mask_rcnn_R_50_FPN_1x_cocostyle.yaml ├── e2e_mask_rcnn_R_50_C4_1x.yaml ├── quick_schedules │ ├── rpn_R_50_C4_quick.yaml │ ├── e2e_faster_rcnn_R_50_C4_quick.yaml │ ├── e2e_mask_rcnn_R_50_C4_quick.yaml │ ├── rpn_R_50_FPN_quick.yaml │ ├── e2e_faster_rcnn_R_50_FPN_quick.yaml │ ├── e2e_faster_rcnn_X_101_32x8d_FPN_quick.yaml │ ├── e2e_mask_rcnn_R_50_FPN_quick.yaml │ ├── e2e_keypoint_rcnn_R_50_FPN_quick.yaml │ └── e2e_mask_rcnn_X_101_32x8d_FPN_quick.yaml ├── rpn_R_50_FPN_1x.yaml ├── rpn_R_101_FPN_1x.yaml ├── rpn_X_101_32x8d_FPN_1x.yaml ├── gn_baselines │ ├── README.md │ ├── e2e_faster_rcnn_R_50_FPN_1x_gn.yaml │ ├── scratch_e2e_faster_rcnn_R_50_FPN_3x_gn.yaml │ ├── e2e_faster_rcnn_R_50_FPN_Xconv1fc_1x_gn.yaml │ ├── scratch_e2e_faster_rcnn_R_50_FPN_Xconv1fc_3x_gn.yaml │ ├── e2e_mask_rcnn_R_50_FPN_1x_gn.yaml │ ├── scratch_e2e_mask_rcnn_R_50_FPN_3x_gn.yaml │ ├── e2e_mask_rcnn_R_50_FPN_Xconv1fc_1x_gn.yaml │ └── scratch_e2e_mask_rcnn_R_50_FPN_Xconv1fc_3x_gn.yaml ├── e2e_faster_rcnn_R_50_FPN_1x.yaml ├── e2e_faster_rcnn_R_101_FPN_1x.yaml ├── cityscapes │ ├── e2e_faster_rcnn_R_50_FPN_1x_cocostyle.yaml │ └── e2e_mask_rcnn_R_50_FPN_1x_cocostyle.yaml ├── e2e_faster_rcnn_X_101_32x8d_FPN_1x.yaml ├── e2e_faster_rcnn_fbnet.yaml ├── e2e_faster_rcnn_fbnet_600.yaml ├── e2e_faster_rcnn_fbnet_chamv1a_600.yaml ├── e2e_mask_rcnn_R_50_FPN_1x.yaml ├── e2e_mask_rcnn_R_101_FPN_1x.yaml ├── retinanet │ ├── retinanet_R-50-FPN_1x_quick.yaml │ ├── retinanet_R-50-FPN_1x.yaml │ ├── retinanet_R-101-FPN_1x.yaml │ ├── retinanet_R-50-FPN_P5_1x.yaml │ ├── retinanet_R-101-FPN_P5_1x.yaml │ └── retinanet_X_101_32x8d_FPN_1x.yaml ├── e2e_keypoint_rcnn_R_50_FPN_1x.yaml ├── e2e_mask_rcnn_X_101_32x8d_FPN_1x.yaml ├── e2e_mask_rcnn_fbnet.yaml ├── e2e_mask_rcnn_fbnet_600.yaml ├── e2e_mask_rcnn_fbnet_xirb16d_dsmask.yaml ├── e2e_mask_rcnn_fbnet_xirb16d_dsmask_600.yaml └── BBTP │ ├── e2e_mask_rcnn_R_101_FPN_4x_voc_aug_cocostyle.yaml │ └── e2e_mask_rcnn_R_101_FPN_4x_voc_coco_aug_cocostyle.yaml ├── test_voc_aug.sh ├── train_voc_aug.sh ├── test_voc_coco_aug.sh ├── CODE_OF_CONDUCT.md ├── train_voc_coco_aug.sh ├── .flake8 ├── tests ├── env_tests │ └── env.py ├── test_configs.py ├── test_metric_logger.py └── utils.py ├── .gitignore ├── docker ├── docker-jupyter │ └── jupyter_notebook_config.py └── Dockerfile ├── LICENSE └── CONTRIBUTING.md /maskrcnn_benchmark/utils/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /maskrcnn_benchmark/modeling/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /maskrcnn_benchmark/structures/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /maskrcnn_benchmark/modeling/roi_heads/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /maskrcnn_benchmark/modeling/rpn/retinanet/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /matlab/DenseCRF/.gitignore: -------------------------------------------------------------------------------- 1 | *.mex* 2 | /data/MSRC/* -------------------------------------------------------------------------------- /maskrcnn_benchmark/modeling/roi_heads/box_head/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/external/liblbfgs/NEWS: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /maskrcnn_benchmark/modeling/roi_heads/keypoint_head/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /maskrcnn_benchmark/modeling/roi_heads/mask_head/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/cppmatrix.cpp: -------------------------------------------------------------------------------- 1 | 2 | 3 | // Nothing 4 | 5 | -------------------------------------------------------------------------------- /requirements.txt: -------------------------------------------------------------------------------- 1 | ninja 2 | yacs 3 | cython 4 | matplotlib 5 | tqdm 6 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/src/SparseCore/SparseAssign.h: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /fig/intro.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chengchunhsu/WSIS_BBTP/HEAD/fig/intro.png -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/Eigen: -------------------------------------------------------------------------------- 1 | #include "Dense" 2 | //#include "Sparse" 3 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/external/liblbfgs/AUTHORS: -------------------------------------------------------------------------------- 1 | Naoaki Okazaki 2 | -------------------------------------------------------------------------------- /maskrcnn_benchmark/__init__.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. 2 | -------------------------------------------------------------------------------- /maskrcnn_benchmark/engine/__init__.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. 2 | -------------------------------------------------------------------------------- /matlab/DenseCRF/examples/img_1001.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chengchunhsu/WSIS_BBTP/HEAD/matlab/DenseCRF/examples/img_1001.jpg -------------------------------------------------------------------------------- /demo/demo_e2e_mask_rcnn_R_50_FPN_1x.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chengchunhsu/WSIS_BBTP/HEAD/demo/demo_e2e_mask_rcnn_R_50_FPN_1x.png -------------------------------------------------------------------------------- /matlab/DenseCRF/include/energy_mex.mexa64: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chengchunhsu/WSIS_BBTP/HEAD/matlab/DenseCRF/include/energy_mex.mexa64 -------------------------------------------------------------------------------- /matlab/DenseCRF/screenshot/screenshot.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chengchunhsu/WSIS_BBTP/HEAD/matlab/DenseCRF/screenshot/screenshot.png -------------------------------------------------------------------------------- /matlab/DenseCRF/examples/img_1001_stage1.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chengchunhsu/WSIS_BBTP/HEAD/matlab/DenseCRF/examples/img_1001_stage1.png -------------------------------------------------------------------------------- /matlab/DenseCRF/examples/img_1001_stage2.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chengchunhsu/WSIS_BBTP/HEAD/matlab/DenseCRF/examples/img_1001_stage2.png -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/external/CMakeLists.txt: -------------------------------------------------------------------------------- 1 | include_directories( liblbfgs/include ) 2 | add_library( lbfgs liblbfgs/lib/lbfgs.cpp ) 3 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf_mex.mexa64: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chengchunhsu/WSIS_BBTP/HEAD/matlab/DenseCRF/include/densecrf_mex.mexa64 -------------------------------------------------------------------------------- /demo/demo_e2e_mask_rcnn_X_101_32x8d_FPN_1x.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chengchunhsu/WSIS_BBTP/HEAD/demo/demo_e2e_mask_rcnn_X_101_32x8d_FPN_1x.png -------------------------------------------------------------------------------- /maskrcnn_benchmark/config/__init__.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. 2 | from .defaults import _C as cfg 3 | -------------------------------------------------------------------------------- /maskrcnn_benchmark/data/__init__.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. 2 | from .build import make_data_loader 3 | -------------------------------------------------------------------------------- /maskrcnn_benchmark/modeling/rpn/__init__.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. 2 | # from .rpn import build_rpn 3 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/TRW_S-v1.3/LICENSE.TXT: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chengchunhsu/WSIS_BBTP/HEAD/matlab/DenseCRF/include/TRW_S-v1.3/LICENSE.TXT -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/examples/im1.ppm: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chengchunhsu/WSIS_BBTP/HEAD/matlab/DenseCRF/include/densecrf/examples/im1.ppm -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/examples/im2.ppm: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chengchunhsu/WSIS_BBTP/HEAD/matlab/DenseCRF/include/densecrf/examples/im2.ppm -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/examples/im3.ppm: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chengchunhsu/WSIS_BBTP/HEAD/matlab/DenseCRF/include/densecrf/examples/im3.ppm -------------------------------------------------------------------------------- /matlab/DenseCRF/include/probImage/compress_zip.sh: -------------------------------------------------------------------------------- 1 | #!/bin/bash 2 | DIR=$(mktemp -d) 3 | bash compress_all.sh $1 $DIR 4 | zip -j $2 $DIR/* 5 | rm -rf $DIR 6 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/examples/anno1.ppm: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chengchunhsu/WSIS_BBTP/HEAD/matlab/DenseCRF/include/densecrf/examples/anno1.ppm -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/examples/anno2.ppm: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chengchunhsu/WSIS_BBTP/HEAD/matlab/DenseCRF/include/densecrf/examples/anno2.ppm -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/examples/anno3.ppm: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chengchunhsu/WSIS_BBTP/HEAD/matlab/DenseCRF/include/densecrf/examples/anno3.ppm -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/src/Core/util/NonMPL2.h: -------------------------------------------------------------------------------- 1 | #ifdef EIGEN_MPL2_ONLY 2 | #error Including non-MPL2 code in EIGEN_MPL2_ONLY mode 3 | #endif 4 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/probImage/decompress_zip.sh: -------------------------------------------------------------------------------- 1 | #!/bin/bash 2 | DIR=$(mktemp -d) 3 | unzip -j $1 -d $DIR/ 4 | bash decompress_all.sh $DIR $2 5 | rm -rf $DIR 6 | -------------------------------------------------------------------------------- /maskrcnn_benchmark/modeling/detector/__init__.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. 2 | from .detectors import build_detection_model 3 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/src/Core/arch/CMakeLists.txt: -------------------------------------------------------------------------------- 1 | ADD_SUBDIRECTORY(SSE) 2 | ADD_SUBDIRECTORY(AltiVec) 3 | ADD_SUBDIRECTORY(NEON) 4 | ADD_SUBDIRECTORY(Default) 5 | -------------------------------------------------------------------------------- /maskrcnn_benchmark/modeling/backbone/__init__.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. 2 | from .backbone import build_backbone 3 | from . import fbnet 4 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/Dense: -------------------------------------------------------------------------------- 1 | #include "Core" 2 | #include "LU" 3 | #include "Cholesky" 4 | #include "QR" 5 | #include "SVD" 6 | #include "Geometry" 7 | #include "Eigenvalues" 8 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/src/QR/CMakeLists.txt: -------------------------------------------------------------------------------- 1 | FILE(GLOB Eigen_QR_SRCS "*.h") 2 | 3 | INSTALL(FILES 4 | ${Eigen_QR_SRCS} 5 | DESTINATION ${INCLUDE_INSTALL_DIR}/Eigen/src/QR COMPONENT Devel 6 | ) 7 | -------------------------------------------------------------------------------- /maskrcnn_benchmark/utils/README.md: -------------------------------------------------------------------------------- 1 | # Utility functions 2 | 3 | This folder contain utility functions that are not used in the 4 | core library, but are useful for building models or training 5 | code using the config system. 6 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/src/SVD/CMakeLists.txt: -------------------------------------------------------------------------------- 1 | FILE(GLOB Eigen_SVD_SRCS "*.h") 2 | 3 | INSTALL(FILES 4 | ${Eigen_SVD_SRCS} 5 | DESTINATION ${INCLUDE_INSTALL_DIR}/Eigen/src/SVD COMPONENT Devel 6 | ) 7 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/src/misc/CMakeLists.txt: -------------------------------------------------------------------------------- 1 | FILE(GLOB Eigen_misc_SRCS "*.h") 2 | 3 | INSTALL(FILES 4 | ${Eigen_misc_SRCS} 5 | DESTINATION ${INCLUDE_INSTALL_DIR}/Eigen/src/misc COMPONENT Devel 6 | ) 7 | -------------------------------------------------------------------------------- /configs/caffe2/e2e_faster_rcnn_R_50_C4_1x_caffe2.yaml: -------------------------------------------------------------------------------- 1 | MODEL: 2 | META_ARCHITECTURE: "GeneralizedRCNN" 3 | WEIGHT: "catalog://Caffe2Detectron/COCO/35857197/e2e_faster_rcnn_R-50-C4_1x" 4 | DATASETS: 5 | TEST: ("coco_2014_minival",) 6 | -------------------------------------------------------------------------------- /maskrcnn_benchmark/solver/__init__.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. 2 | from .build import make_optimizer 3 | from .build import make_lr_scheduler 4 | from .lr_scheduler import WarmupMultiStepLR 5 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/src/Jacobi/CMakeLists.txt: -------------------------------------------------------------------------------- 1 | FILE(GLOB Eigen_Jacobi_SRCS "*.h") 2 | 3 | INSTALL(FILES 4 | ${Eigen_Jacobi_SRCS} 5 | DESTINATION ${INCLUDE_INSTALL_DIR}/Eigen/src/Jacobi COMPONENT Devel 6 | ) 7 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/src/LU/arch/CMakeLists.txt: -------------------------------------------------------------------------------- 1 | FILE(GLOB Eigen_LU_arch_SRCS "*.h") 2 | 3 | INSTALL(FILES 4 | ${Eigen_LU_arch_SRCS} 5 | DESTINATION ${INCLUDE_INSTALL_DIR}/Eigen/src/LU/arch COMPONENT Devel 6 | ) 7 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/src/plugins/CMakeLists.txt: -------------------------------------------------------------------------------- 1 | FILE(GLOB Eigen_plugins_SRCS "*.h") 2 | 3 | INSTALL(FILES 4 | ${Eigen_plugins_SRCS} 5 | DESTINATION ${INCLUDE_INSTALL_DIR}/Eigen/src/plugins COMPONENT Devel 6 | ) 7 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/src/Cholesky/CMakeLists.txt: -------------------------------------------------------------------------------- 1 | FILE(GLOB Eigen_Cholesky_SRCS "*.h") 2 | 3 | INSTALL(FILES 4 | ${Eigen_Cholesky_SRCS} 5 | DESTINATION ${INCLUDE_INSTALL_DIR}/Eigen/src/Cholesky COMPONENT Devel 6 | ) 7 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/src/Core/util/CMakeLists.txt: -------------------------------------------------------------------------------- 1 | FILE(GLOB Eigen_Core_util_SRCS "*.h") 2 | 3 | INSTALL(FILES 4 | ${Eigen_Core_util_SRCS} 5 | DESTINATION ${INCLUDE_INSTALL_DIR}/Eigen/src/Core/util COMPONENT Devel 6 | ) 7 | -------------------------------------------------------------------------------- /matlab/DenseCRF/examples/example_MSRC.m: -------------------------------------------------------------------------------- 1 | clear all; close all; 2 | addpath('..'); 3 | 4 | %% 5 | image_name= '2_14_s'; 6 | M = MSRC(image_name); 7 | 8 | M.bilateral_weight =1; 9 | M.gaussian_weight = 1; 10 | 11 | %% 12 | M.mean_field(); 13 | M.display() -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/src/Eigenvalues/CMakeLists.txt: -------------------------------------------------------------------------------- 1 | FILE(GLOB Eigen_EIGENVALUES_SRCS "*.h") 2 | 3 | INSTALL(FILES 4 | ${Eigen_EIGENVALUES_SRCS} 5 | DESTINATION ${INCLUDE_INSTALL_DIR}/Eigen/src/Eigenvalues COMPONENT Devel 6 | ) 7 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/src/Householder/CMakeLists.txt: -------------------------------------------------------------------------------- 1 | FILE(GLOB Eigen_Householder_SRCS "*.h") 2 | 3 | INSTALL(FILES 4 | ${Eigen_Householder_SRCS} 5 | DESTINATION ${INCLUDE_INSTALL_DIR}/Eigen/src/Householder COMPONENT Devel 6 | ) 7 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/src/LU/CMakeLists.txt: -------------------------------------------------------------------------------- 1 | FILE(GLOB Eigen_LU_SRCS "*.h") 2 | 3 | INSTALL(FILES 4 | ${Eigen_LU_SRCS} 5 | DESTINATION ${INCLUDE_INSTALL_DIR}/Eigen/src/LU COMPONENT Devel 6 | ) 7 | 8 | ADD_SUBDIRECTORY(arch) 9 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/src/SparseCore/CMakeLists.txt: -------------------------------------------------------------------------------- 1 | FILE(GLOB Eigen_SparseCore_SRCS "*.h") 2 | 3 | INSTALL(FILES 4 | ${Eigen_SparseCore_SRCS} 5 | DESTINATION ${INCLUDE_INSTALL_DIR}/Eigen/src/SparseCore COMPONENT Devel 6 | ) 7 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/src/StlSupport/CMakeLists.txt: -------------------------------------------------------------------------------- 1 | FILE(GLOB Eigen_StlSupport_SRCS "*.h") 2 | 3 | INSTALL(FILES 4 | ${Eigen_StlSupport_SRCS} 5 | DESTINATION ${INCLUDE_INSTALL_DIR}/Eigen/src/StlSupport COMPONENT Devel 6 | ) 7 | -------------------------------------------------------------------------------- /test_voc_aug.sh: -------------------------------------------------------------------------------- 1 | CUDA_VISIBLE_DEVICES=0,1,2,3 python -m torch.distributed.launch \ 2 | --nproc_per_node=4 \ 3 | --master_port=$((RANDOM + 10000)) \ 4 | tools/test_net.py \ 5 | --config-file ./configs/BBTP/e2e_mask_rcnn_R_101_FPN_4x_voc_aug_cocostyle.yaml -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/src/Core/arch/NEON/CMakeLists.txt: -------------------------------------------------------------------------------- 1 | FILE(GLOB Eigen_Core_arch_NEON_SRCS "*.h") 2 | 3 | INSTALL(FILES 4 | ${Eigen_Core_arch_NEON_SRCS} 5 | DESTINATION ${INCLUDE_INSTALL_DIR}/Eigen/src/Core/arch/NEON COMPONENT Devel 6 | ) 7 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/src/Core/arch/SSE/CMakeLists.txt: -------------------------------------------------------------------------------- 1 | FILE(GLOB Eigen_Core_arch_SSE_SRCS "*.h") 2 | 3 | INSTALL(FILES 4 | ${Eigen_Core_arch_SSE_SRCS} 5 | DESTINATION ${INCLUDE_INSTALL_DIR}/Eigen/src/Core/arch/SSE COMPONENT Devel 6 | ) 7 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/src/Core/products/CMakeLists.txt: -------------------------------------------------------------------------------- 1 | FILE(GLOB Eigen_Core_Product_SRCS "*.h") 2 | 3 | INSTALL(FILES 4 | ${Eigen_Core_Product_SRCS} 5 | DESTINATION ${INCLUDE_INSTALL_DIR}/Eigen/src/Core/products COMPONENT Devel 6 | ) 7 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/src/Geometry/arch/CMakeLists.txt: -------------------------------------------------------------------------------- 1 | FILE(GLOB Eigen_Geometry_arch_SRCS "*.h") 2 | 3 | INSTALL(FILES 4 | ${Eigen_Geometry_arch_SRCS} 5 | DESTINATION ${INCLUDE_INSTALL_DIR}/Eigen/src/Geometry/arch COMPONENT Devel 6 | ) 7 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/src/PaStiXSupport/CMakeLists.txt: -------------------------------------------------------------------------------- 1 | FILE(GLOB Eigen_PastixSupport_SRCS "*.h") 2 | 3 | INSTALL(FILES 4 | ${Eigen_PastixSupport_SRCS} 5 | DESTINATION ${INCLUDE_INSTALL_DIR}/Eigen/src/PaStiXSupport COMPONENT Devel 6 | ) 7 | -------------------------------------------------------------------------------- /train_voc_aug.sh: -------------------------------------------------------------------------------- 1 | CUDA_VISIBLE_DEVICES=0,1,2,3 python -m torch.distributed.launch \ 2 | --nproc_per_node=4 \ 3 | --master_port=$((RANDOM + 10000)) \ 4 | tools/train_net.py \ 5 | --config-file ./configs/BBTP/e2e_mask_rcnn_R_101_FPN_4x_voc_aug_cocostyle.yaml -------------------------------------------------------------------------------- /maskrcnn_benchmark/layers/nms.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. 2 | # from ._utils import _C 3 | from maskrcnn_benchmark import _C 4 | 5 | nms = _C.nms 6 | # nms.__doc__ = """ 7 | # This function performs Non-maximum suppresion""" 8 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/src/CholmodSupport/CMakeLists.txt: -------------------------------------------------------------------------------- 1 | FILE(GLOB Eigen_CholmodSupport_SRCS "*.h") 2 | 3 | INSTALL(FILES 4 | ${Eigen_CholmodSupport_SRCS} 5 | DESTINATION ${INCLUDE_INSTALL_DIR}/Eigen/src/CholmodSupport COMPONENT Devel 6 | ) 7 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/src/PardisoSupport/CMakeLists.txt: -------------------------------------------------------------------------------- 1 | FILE(GLOB Eigen_PardisoSupport_SRCS "*.h") 2 | 3 | INSTALL(FILES 4 | ${Eigen_PardisoSupport_SRCS} 5 | DESTINATION ${INCLUDE_INSTALL_DIR}/Eigen/src/PardisoSupport COMPONENT Devel 6 | ) 7 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/src/SparseCholesky/CMakeLists.txt: -------------------------------------------------------------------------------- 1 | FILE(GLOB Eigen_SparseCholesky_SRCS "*.h") 2 | 3 | INSTALL(FILES 4 | ${Eigen_SparseCholesky_SRCS} 5 | DESTINATION ${INCLUDE_INSTALL_DIR}/Eigen/src/SparseCholesky COMPONENT Devel 6 | ) 7 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/src/SuperLUSupport/CMakeLists.txt: -------------------------------------------------------------------------------- 1 | FILE(GLOB Eigen_SuperLUSupport_SRCS "*.h") 2 | 3 | INSTALL(FILES 4 | ${Eigen_SuperLUSupport_SRCS} 5 | DESTINATION ${INCLUDE_INSTALL_DIR}/Eigen/src/SuperLUSupport COMPONENT Devel 6 | ) 7 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/src/UmfPackSupport/CMakeLists.txt: -------------------------------------------------------------------------------- 1 | FILE(GLOB Eigen_UmfPackSupport_SRCS "*.h") 2 | 3 | INSTALL(FILES 4 | ${Eigen_UmfPackSupport_SRCS} 5 | DESTINATION ${INCLUDE_INSTALL_DIR}/Eigen/src/UmfPackSupport COMPONENT Devel 6 | ) 7 | -------------------------------------------------------------------------------- /test_voc_coco_aug.sh: -------------------------------------------------------------------------------- 1 | CUDA_VISIBLE_DEVICES=0,1,2,3 python -m torch.distributed.launch \ 2 | --nproc_per_node=4 \ 3 | --master_port=$((RANDOM + 10000)) \ 4 | tools/test_net.py \ 5 | --config-file ./configs/BBTP/e2e_mask_rcnn_R_101_FPN_4x_voc_coco_aug_cocostyle.yaml -------------------------------------------------------------------------------- /CODE_OF_CONDUCT.md: -------------------------------------------------------------------------------- 1 | # Code of Conduct 2 | 3 | Facebook has adopted a Code of Conduct that we expect project participants to adhere to. 4 | Please read the [full text](https://code.fb.com/codeofconduct/) 5 | so that you can understand what actions will and will not be tolerated. 6 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/external/liblbfgs/Makefile.am: -------------------------------------------------------------------------------- 1 | # $Id$ 2 | 3 | SUBDIRS = lib sample 4 | 5 | docdir = $(prefix)/share/doc/@PACKAGE@ 6 | doc_DATA = README INSTALL COPYING AUTHORS ChangeLog NEWS 7 | 8 | EXTRA_DIST = \ 9 | autogen.sh \ 10 | lbfgs.sln 11 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/src/Geometry/CMakeLists.txt: -------------------------------------------------------------------------------- 1 | FILE(GLOB Eigen_Geometry_SRCS "*.h") 2 | 3 | INSTALL(FILES 4 | ${Eigen_Geometry_SRCS} 5 | DESTINATION ${INCLUDE_INSTALL_DIR}/Eigen/src/Geometry COMPONENT Devel 6 | ) 7 | 8 | ADD_SUBDIRECTORY(arch) 9 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/src/OrderingMethods/CMakeLists.txt: -------------------------------------------------------------------------------- 1 | FILE(GLOB Eigen_OrderingMethods_SRCS "*.h") 2 | 3 | INSTALL(FILES 4 | ${Eigen_OrderingMethods_SRCS} 5 | DESTINATION ${INCLUDE_INSTALL_DIR}/Eigen/src/OrderingMethods COMPONENT Devel 6 | ) 7 | -------------------------------------------------------------------------------- /train_voc_coco_aug.sh: -------------------------------------------------------------------------------- 1 | CUDA_VISIBLE_DEVICES=0,1,2,3 python -m torch.distributed.launch \ 2 | --nproc_per_node=4 \ 3 | --master_port=$((RANDOM + 10000)) \ 4 | tools/train_net.py \ 5 | --config-file ./configs/BBTP/e2e_mask_rcnn_R_101_FPN_4x_voc_coco_aug_cocostyle.yaml -------------------------------------------------------------------------------- /.flake8: -------------------------------------------------------------------------------- 1 | # This is an example .flake8 config, used when developing *Black* itself. 2 | # Keep in sync with setup.cfg which is used for source packages. 3 | 4 | [flake8] 5 | ignore = E203, E266, E501, W503 6 | max-line-length = 80 7 | max-complexity = 18 8 | select = B,C,E,F,W,T4,B9 9 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/src/Core/arch/AltiVec/CMakeLists.txt: -------------------------------------------------------------------------------- 1 | FILE(GLOB Eigen_Core_arch_AltiVec_SRCS "*.h") 2 | 3 | INSTALL(FILES 4 | ${Eigen_Core_arch_AltiVec_SRCS} 5 | DESTINATION ${INCLUDE_INSTALL_DIR}/Eigen/src/Core/arch/AltiVec COMPONENT Devel 6 | ) 7 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/src/Core/arch/Default/CMakeLists.txt: -------------------------------------------------------------------------------- 1 | FILE(GLOB Eigen_Core_arch_Default_SRCS "*.h") 2 | 3 | INSTALL(FILES 4 | ${Eigen_Core_arch_Default_SRCS} 5 | DESTINATION ${INCLUDE_INSTALL_DIR}/Eigen/src/Core/arch/Default COMPONENT Devel 6 | ) 7 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/src/Eigen2Support/Geometry/CMakeLists.txt: -------------------------------------------------------------------------------- 1 | FILE(GLOB Eigen_Eigen2Support_Geometry_SRCS "*.h") 2 | 3 | INSTALL(FILES 4 | ${Eigen_Eigen2Support_Geometry_SRCS} 5 | DESTINATION ${INCLUDE_INSTALL_DIR}/Eigen/src/Eigen2Support/Geometry 6 | ) 7 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/src/Eigen2Support/CMakeLists.txt: -------------------------------------------------------------------------------- 1 | FILE(GLOB Eigen_Eigen2Support_SRCS "*.h") 2 | 3 | INSTALL(FILES 4 | ${Eigen_Eigen2Support_SRCS} 5 | DESTINATION ${INCLUDE_INSTALL_DIR}/Eigen/src/Eigen2Support COMPONENT Devel 6 | ) 7 | 8 | ADD_SUBDIRECTORY(Geometry) -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/src/IterativeLinearSolvers/CMakeLists.txt: -------------------------------------------------------------------------------- 1 | FILE(GLOB Eigen_IterativeLinearSolvers_SRCS "*.h") 2 | 3 | INSTALL(FILES 4 | ${Eigen_IterativeLinearSolvers_SRCS} 5 | DESTINATION ${INCLUDE_INSTALL_DIR}/Eigen/src/IterativeLinearSolvers COMPONENT Devel 6 | ) 7 | -------------------------------------------------------------------------------- /maskrcnn_benchmark/data/datasets/__init__.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. 2 | from .coco import COCODataset 3 | from .voc import PascalVOCDataset 4 | from .concat_dataset import ConcatDataset 5 | 6 | __all__ = ["COCODataset", "ConcatDataset", "PascalVOCDataset"] 7 | -------------------------------------------------------------------------------- /maskrcnn_benchmark/utils/miscellaneous.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. 2 | import errno 3 | import os 4 | 5 | 6 | def mkdir(path): 7 | try: 8 | os.makedirs(path) 9 | except OSError as e: 10 | if e.errno != errno.EEXIST: 11 | raise 12 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/external/liblbfgs/sample/Makefile.am: -------------------------------------------------------------------------------- 1 | # $Id$ 2 | 3 | EXTRA_DIST = \ 4 | sample.cpp \ 5 | sample.vcxproj 6 | 7 | noinst_PROGRAMS = sample 8 | 9 | sample_SOURCES = \ 10 | sample.c 11 | 12 | sample_LDADD = ../lib/liblbfgs.la 13 | 14 | AM_CFLAGS = @CFLAGS@ 15 | INCLUDES = @INCLUDES@ 16 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/CMakeLists.txt: -------------------------------------------------------------------------------- 1 | cmake_minimum_required (VERSION 2.6) 2 | project(densecrf) 3 | 4 | set(CMAKE_MODULE_PATH ${CMAKE_MODULE_PATH} ${PROJECT_SOURCE_DIR}/cmake) 5 | 6 | include_directories( include ) 7 | 8 | add_subdirectory (src) 9 | add_subdirectory (examples) 10 | add_subdirectory (external) 11 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/src/Core/CMakeLists.txt: -------------------------------------------------------------------------------- 1 | FILE(GLOB Eigen_Core_SRCS "*.h") 2 | 3 | INSTALL(FILES 4 | ${Eigen_Core_SRCS} 5 | DESTINATION ${INCLUDE_INSTALL_DIR}/Eigen/src/Core COMPONENT Devel 6 | ) 7 | 8 | ADD_SUBDIRECTORY(products) 9 | ADD_SUBDIRECTORY(util) 10 | ADD_SUBDIRECTORY(arch) 11 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/src/CMakeLists.txt: -------------------------------------------------------------------------------- 1 | include_directories( ../external/liblbfgs/include ../external/ ) 2 | 3 | add_library( densecrf util.cpp permutohedral.cpp unary.cpp pairwise.cpp objective.cpp densecrf.cpp labelcompatibility.cpp ) 4 | 5 | add_library( optimization optimization.cpp ) 6 | target_link_libraries( optimization lbfgs ) 7 | -------------------------------------------------------------------------------- /maskrcnn_benchmark/data/transforms/__init__.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. 2 | from .transforms import Compose 3 | from .transforms import Resize 4 | from .transforms import RandomHorizontalFlip 5 | from .transforms import ToTensor 6 | from .transforms import Normalize 7 | 8 | from .build import build_transforms 9 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/probImage/compress_all.sh: -------------------------------------------------------------------------------- 1 | #!/bin/bash 2 | EXE=build/compress 3 | 4 | if [ $# -lt 2 ]; then 5 | echo "Usage: $0 unary_dir compressed_dir" 6 | exit 1 7 | fi 8 | 9 | for i in $1/*.unary; do 10 | PATH=$i 11 | NAME=${i/$1\//} 12 | TARGET=${NAME/unary/c_unary} 13 | TARGET_PATH=$2/$TARGET 14 | $EXE $PATH $TARGET_PATH 15 | done 16 | -------------------------------------------------------------------------------- /configs/caffe2/e2e_mask_rcnn_R_50_C4_1x_caffe2.yaml: -------------------------------------------------------------------------------- 1 | MODEL: 2 | META_ARCHITECTURE: "GeneralizedRCNN" 3 | WEIGHT: "catalog://Caffe2Detectron/COCO/35858791/e2e_mask_rcnn_R-50-C4_1x" 4 | ROI_MASK_HEAD: 5 | PREDICTOR: "MaskRCNNC4Predictor" 6 | SHARE_BOX_FEATURE_EXTRACTOR: True 7 | MASK_ON: True 8 | DATASETS: 9 | TEST: ("coco_2014_minival",) 10 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/probImage/CMakeLists.txt: -------------------------------------------------------------------------------- 1 | cmake_minimum_required (VERSION 2.6) 2 | project(probimage) 3 | 4 | add_library( probimage probimage.cpp ) 5 | 6 | add_executable( compress compress.cpp ) 7 | target_link_libraries( compress probimage ) 8 | 9 | add_executable( decompress decompress.cpp ) 10 | target_link_libraries( decompress probimage ) 11 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/probImage/compress.cpp: -------------------------------------------------------------------------------- 1 | #include "probimage.h" 2 | #include 3 | 4 | int main( int argc, char * argv[] ){ 5 | if (argc<3){ 6 | printf("Usage: %s file compressed_file\n", argv[0] ); 7 | } 8 | ProbImage im; 9 | im.load( argv[1] ); 10 | im.boostToProb(); 11 | im.compress( argv[2], 0.001 ); 12 | 13 | return 0; 14 | } 15 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/probImage/decompress.cpp: -------------------------------------------------------------------------------- 1 | #include "probimage.h" 2 | #include 3 | 4 | int main( int argc, char * argv[] ){ 5 | if (argc<3){ 6 | printf("Usage: %s compressed_file file\n", argv[0] ); 7 | } 8 | ProbImage im; 9 | im.decompress( argv[1] ); 10 | // im.probToEnergy(); 11 | im.save( argv[2] ); 12 | 13 | return 0; 14 | } 15 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/probImage/decompress_all.sh: -------------------------------------------------------------------------------- 1 | #!/bin/bash 2 | EXE=build/decompress 3 | 4 | if [ $# -lt 2 ]; then 5 | echo "Usage: $0 compressed_dir unary_dir" 6 | exit 1 7 | fi 8 | 9 | for i in $1/*.c_unary; do 10 | PATH=$i 11 | NAME=${i/$1\//} 12 | TARGET=${NAME/c_unary/unary} 13 | TARGET_PATH=$2/$TARGET 14 | $EXE $PATH $TARGET_PATH 15 | done 16 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/solvers.h: -------------------------------------------------------------------------------- 1 | #pragma once 2 | 3 | #include "TRW_S-v1.3/MRFEnergy.h" 4 | typedef MRFEnergy TRWS; 5 | typedef MRFEnergy::NodeId TRWSNodes; 6 | typedef MRFEnergy::Options TRWSOptions; 7 | 8 | #include "maxflow-v3.03.src/graph.h" 9 | typedef Graph GraphType; 10 | 11 | #include "extendedDenseCRF2D.h" -------------------------------------------------------------------------------- /maskrcnn_benchmark/data/samplers/__init__.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. 2 | from .distributed import DistributedSampler 3 | from .grouped_batch_sampler import GroupedBatchSampler 4 | from .iteration_based_batch_sampler import IterationBasedBatchSampler 5 | 6 | __all__ = ["DistributedSampler", "GroupedBatchSampler", "IterationBasedBatchSampler"] 7 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/src/CMakeLists.txt: -------------------------------------------------------------------------------- 1 | file(GLOB Eigen_src_subdirectories "*") 2 | escape_string_as_regex(ESCAPED_CMAKE_CURRENT_SOURCE_DIR "${CMAKE_CURRENT_SOURCE_DIR}") 3 | foreach(f ${Eigen_src_subdirectories}) 4 | if(NOT f MATCHES "\\.txt" AND NOT f MATCHES "${ESCAPED_CMAKE_CURRENT_SOURCE_DIR}/[.].+" ) 5 | add_subdirectory(${f}) 6 | endif() 7 | endforeach() 8 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/TRW_S-v1.3/CHANGES.TXT: -------------------------------------------------------------------------------- 1 | Changes from version 1.2: 2 | 3 | - Fixed bug in typeBinaryFast 4 | - Added the option of reading min_marginals 5 | - Added function AddRandomMessages() 6 | 7 | Changes from version 1.1: 8 | 9 | - Fixed bug in memory allocation 10 | 11 | Changes from version 1.0: 12 | 13 | - Modified syntax of 'friend' declarations to make it compile under unix -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/Array: -------------------------------------------------------------------------------- 1 | #ifndef EIGEN_ARRAY_MODULE_H 2 | #define EIGEN_ARRAY_MODULE_H 3 | 4 | // include Core first to handle Eigen2 support macros 5 | #include "Core" 6 | 7 | #ifndef EIGEN2_SUPPORT 8 | #error The Eigen/Array header does no longer exist in Eigen3. All that functionality has moved to Eigen/Core. 9 | #endif 10 | 11 | #endif // EIGEN_ARRAY_MODULE_H 12 | -------------------------------------------------------------------------------- /maskrcnn_benchmark/modeling/detector/detectors.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. 2 | from .generalized_rcnn import GeneralizedRCNN 3 | 4 | 5 | _DETECTION_META_ARCHITECTURES = {"GeneralizedRCNN": GeneralizedRCNN} 6 | 7 | 8 | def build_detection_model(cfg): 9 | meta_arch = _DETECTION_META_ARCHITECTURES[cfg.MODEL.META_ARCHITECTURE] 10 | return meta_arch(cfg) 11 | -------------------------------------------------------------------------------- /maskrcnn_benchmark/utils/collect_env.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. 2 | import PIL 3 | 4 | from torch.utils.collect_env import get_pretty_env_info 5 | 6 | 7 | def get_pil_version(): 8 | return "\n Pillow ({})".format(PIL.__version__) 9 | 10 | 11 | def collect_env_info(): 12 | env_str = get_pretty_env_info() 13 | env_str += get_pil_version() 14 | return env_str 15 | -------------------------------------------------------------------------------- /configs/rpn_R_50_C4_1x.yaml: -------------------------------------------------------------------------------- 1 | MODEL: 2 | META_ARCHITECTURE: "GeneralizedRCNN" 3 | WEIGHT: "catalog://ImageNetPretrained/MSRA/R-50" 4 | RPN_ONLY: True 5 | RPN: 6 | PRE_NMS_TOP_N_TEST: 12000 7 | POST_NMS_TOP_N_TEST: 2000 8 | DATASETS: 9 | TRAIN: ("coco_2014_train", "coco_2014_valminusminival") 10 | TEST: ("coco_2014_minival",) 11 | SOLVER: 12 | BASE_LR: 0.02 13 | WEIGHT_DECAY: 0.0001 14 | STEPS: (60000, 80000) 15 | MAX_ITER: 90000 16 | -------------------------------------------------------------------------------- /tests/env_tests/env.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. 2 | 3 | import os 4 | 5 | 6 | def get_config_root_path(): 7 | ''' Path to configs for unit tests ''' 8 | # cur_file_dir is root/tests/env_tests 9 | cur_file_dir = os.path.dirname(os.path.abspath(os.path.realpath(__file__))) 10 | ret = os.path.dirname(os.path.dirname(cur_file_dir)) 11 | ret = os.path.join(ret, "configs") 12 | return ret 13 | -------------------------------------------------------------------------------- /.gitignore: -------------------------------------------------------------------------------- 1 | # compilation and distribution 2 | __pycache__ 3 | _ext 4 | *.pyc 5 | *.so 6 | maskrcnn_benchmark.egg-info/ 7 | build/ 8 | dist/ 9 | 10 | # pytorch/python/numpy formats 11 | *.pth 12 | *.pkl 13 | *.npy 14 | 15 | # ipython/jupyter notebooks 16 | *.ipynb 17 | **/.ipynb_checkpoints/ 18 | 19 | # Editor temporaries 20 | *.swn 21 | *.swo 22 | *.swp 23 | *~ 24 | 25 | # Pycharm editor settings 26 | .idea 27 | 28 | # project dirs 29 | /datasets 30 | /models 31 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/TRW_S-v1.3/instances.inc: -------------------------------------------------------------------------------- 1 | 2 | #include "MRFEnergy.h" 3 | 4 | template class MRFEnergy; 5 | template class MRFEnergy; 6 | template class MRFEnergy; 7 | template class MRFEnergy; 8 | template class MRFEnergy; 9 | template class MRFEnergy; 10 | template class MRFEnergy; 11 | template class MRFEnergy; 12 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/src/Core/util/ReenableStupidWarnings.h: -------------------------------------------------------------------------------- 1 | #ifdef EIGEN_WARNINGS_DISABLED 2 | #undef EIGEN_WARNINGS_DISABLED 3 | 4 | #ifndef EIGEN_PERMANENTLY_DISABLE_STUPID_WARNINGS 5 | #ifdef _MSC_VER 6 | #pragma warning( pop ) 7 | #elif defined __INTEL_COMPILER 8 | #pragma warning pop 9 | #elif defined __clang__ 10 | #pragma clang diagnostic pop 11 | #endif 12 | #endif 13 | 14 | #endif // EIGEN_WARNINGS_DISABLED 15 | -------------------------------------------------------------------------------- /configs/e2e_faster_rcnn_R_50_C4_1x.yaml: -------------------------------------------------------------------------------- 1 | MODEL: 2 | META_ARCHITECTURE: "GeneralizedRCNN" 3 | WEIGHT: "catalog://ImageNetPretrained/MSRA/R-50" 4 | RPN: 5 | PRE_NMS_TOP_N_TEST: 6000 6 | POST_NMS_TOP_N_TEST: 1000 7 | DATASETS: 8 | TRAIN: ("coco_2014_train", "coco_2014_valminusminival") 9 | TEST: ("coco_2014_minival",) 10 | SOLVER: 11 | BASE_LR: 0.01 12 | WEIGHT_DECAY: 0.0001 13 | STEPS: (120000, 160000) 14 | MAX_ITER: 180000 15 | IMS_PER_BATCH: 8 16 | -------------------------------------------------------------------------------- /maskrcnn_benchmark/modeling/registry.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. 2 | 3 | from maskrcnn_benchmark.utils.registry import Registry 4 | 5 | BACKBONES = Registry() 6 | RPN_HEADS = Registry() 7 | ROI_BOX_FEATURE_EXTRACTORS = Registry() 8 | ROI_BOX_PREDICTOR = Registry() 9 | ROI_KEYPOINT_FEATURE_EXTRACTORS = Registry() 10 | ROI_KEYPOINT_PREDICTOR = Registry() 11 | ROI_MASK_FEATURE_EXTRACTORS = Registry() 12 | ROI_MASK_PREDICTOR = Registry() 13 | -------------------------------------------------------------------------------- /maskrcnn_benchmark/modeling/utils.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. 2 | """ 3 | Miscellaneous utility functions 4 | """ 5 | 6 | import torch 7 | 8 | 9 | def cat(tensors, dim=0): 10 | """ 11 | Efficient version of torch.cat that avoids a copy if there is only a single element in a list 12 | """ 13 | assert isinstance(tensors, (list, tuple)) 14 | if len(tensors) == 1: 15 | return tensors[0] 16 | return torch.cat(tensors, dim) 17 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/maxflow-v3.03.src/instances.inc: -------------------------------------------------------------------------------- 1 | #include "graph.h" 2 | 3 | #ifdef _MSC_VER 4 | #pragma warning(disable: 4661) 5 | #endif 6 | 7 | // Instantiations: 8 | // IMPORTANT: 9 | // flowtype should be 'larger' than tcaptype 10 | // tcaptype should be 'larger' than captype 11 | 12 | template class Graph; 13 | template class Graph; 14 | template class Graph; 15 | template class Graph; 16 | 17 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/external/liblbfgs/lib/Makefile.am: -------------------------------------------------------------------------------- 1 | # $Id$ 2 | 3 | EXTRA_DIST = \ 4 | lib.vcxproj 5 | 6 | liblbfgsincludedir = $(includedir) 7 | liblbfgsinclude_HEADERS = \ 8 | ../include/lbfgs.h 9 | 10 | lib_LTLIBRARIES = liblbfgs.la 11 | 12 | liblbfgs_la_SOURCES = \ 13 | arithmetic_ansi.h \ 14 | arithmetic_sse_double.h \ 15 | arithmetic_sse_float.h \ 16 | ../include/lbfgs.h \ 17 | lbfgs.c 18 | 19 | liblbfgs_la_LDFLAGS = \ 20 | -no-undefined \ 21 | -release @VERSION@ 22 | 23 | AM_CFLAGS = @CFLAGS@ 24 | INCLUDES = @INCLUDES@ 25 | -------------------------------------------------------------------------------- /matlab/DenseCRF/data/readme.txt: -------------------------------------------------------------------------------- 1 | In order to use the MSRC class you need to download and unpack 2 | 1. Unary potentials from http://graphics.stanford.edu/projects/densecrf/unary/ 3 | 2. MSRC-21 dataset from http://research.microsoft.com/en-us/projects/objectclassrecognition/ 4 | 3. More accurate annotations http://graphics.stanford.edu/projects/densecrf/textonboost/ 5 | 6 | The code assumes the catalog structure to be: 7 | /data/MSRC/HighQuality 8 | /data/MSRC/Images 9 | /data/MSRC/SegmentationsGTHighQuality 10 | /data/MSRC/split 11 | /data/MSRC/unary -------------------------------------------------------------------------------- /configs/pascal_voc/e2e_faster_rcnn_R_50_C4_1x_1_gpu_voc.yaml: -------------------------------------------------------------------------------- 1 | MODEL: 2 | META_ARCHITECTURE: "GeneralizedRCNN" 3 | WEIGHT: "catalog://ImageNetPretrained/MSRA/R-50" 4 | RPN: 5 | PRE_NMS_TOP_N_TEST: 6000 6 | POST_NMS_TOP_N_TEST: 300 7 | ANCHOR_SIZES: (128, 256, 512) 8 | ROI_BOX_HEAD: 9 | NUM_CLASSES: 21 10 | DATASETS: 11 | TRAIN: ("voc_2007_train", "voc_2007_val") 12 | TEST: ("voc_2007_test",) 13 | SOLVER: 14 | BASE_LR: 0.001 15 | WEIGHT_DECAY: 0.0001 16 | STEPS: (50000, ) 17 | MAX_ITER: 70000 18 | IMS_PER_BATCH: 1 19 | TEST: 20 | IMS_PER_BATCH: 1 21 | -------------------------------------------------------------------------------- /configs/pascal_voc/e2e_faster_rcnn_R_50_C4_1x_4_gpu_voc.yaml: -------------------------------------------------------------------------------- 1 | MODEL: 2 | META_ARCHITECTURE: "GeneralizedRCNN" 3 | WEIGHT: "catalog://ImageNetPretrained/MSRA/R-50" 4 | RPN: 5 | PRE_NMS_TOP_N_TEST: 6000 6 | POST_NMS_TOP_N_TEST: 300 7 | ANCHOR_SIZES: (128, 256, 512) 8 | ROI_BOX_HEAD: 9 | NUM_CLASSES: 21 10 | DATASETS: 11 | TRAIN: ("voc_2007_train", "voc_2007_val") 12 | TEST: ("voc_2007_test",) 13 | SOLVER: 14 | BASE_LR: 0.004 15 | WEIGHT_DECAY: 0.0001 16 | STEPS: (12500, ) 17 | MAX_ITER: 17500 18 | IMS_PER_BATCH: 4 19 | TEST: 20 | IMS_PER_BATCH: 4 21 | -------------------------------------------------------------------------------- /maskrcnn_benchmark/layers/smooth_l1_loss.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. 2 | import torch 3 | 4 | 5 | # TODO maybe push this to nn? 6 | def smooth_l1_loss(input, target, beta=1. / 9, size_average=True): 7 | """ 8 | very similar to the smooth_l1_loss from pytorch, but with 9 | the extra beta parameter 10 | """ 11 | n = torch.abs(input - target) 12 | cond = n < beta 13 | loss = torch.where(cond, 0.5 * n ** 2 / beta, n - 0.5 * beta) 14 | if size_average: 15 | return loss.mean() 16 | return loss.sum() 17 | -------------------------------------------------------------------------------- /configs/e2e_mask_rcnn_R_50_C4_1x.yaml: -------------------------------------------------------------------------------- 1 | MODEL: 2 | META_ARCHITECTURE: "GeneralizedRCNN" 3 | WEIGHT: "catalog://ImageNetPretrained/MSRA/R-50" 4 | RPN: 5 | PRE_NMS_TOP_N_TEST: 6000 6 | POST_NMS_TOP_N_TEST: 1000 7 | ROI_MASK_HEAD: 8 | PREDICTOR: "MaskRCNNC4Predictor" 9 | SHARE_BOX_FEATURE_EXTRACTOR: True 10 | MASK_ON: True 11 | DATASETS: 12 | TRAIN: ("coco_2014_train", "coco_2014_valminusminival") 13 | TEST: ("coco_2014_minival",) 14 | SOLVER: 15 | BASE_LR: 0.01 16 | WEIGHT_DECAY: 0.0001 17 | STEPS: (120000, 160000) 18 | MAX_ITER: 180000 19 | IMS_PER_BATCH: 8 20 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/TRW_S-v1.3/instances.h: -------------------------------------------------------------------------------- 1 | #ifndef __INSTANCES_H__ 2 | #define __INSTANCES_H__ 3 | 4 | 5 | #if defined(_MSC_VER) 6 | 7 | // C4661: '...' : no suitable definition provided for explicit template instantiation request 8 | #pragma warning(disable: 4661) 9 | 10 | #endif 11 | 12 | #include "typeBinary.h" 13 | #include "typeBinaryFast.h" 14 | #include "typePotts.h" 15 | #include "typeGeneral.h" 16 | #include "typeTruncatedLinear.h" 17 | #include "typeTruncatedQuadratic.h" 18 | #include "typeTruncatedLinear2D.h" 19 | #include "typeTruncatedQuadratic2D.h" 20 | 21 | 22 | #endif 23 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/Sparse: -------------------------------------------------------------------------------- 1 | #ifndef EIGEN_SPARSE_MODULE_H 2 | #define EIGEN_SPARSE_MODULE_H 3 | 4 | /** \defgroup Sparse_modules Sparse modules 5 | * 6 | * Meta-module including all related modules: 7 | * - SparseCore 8 | * - OrderingMethods 9 | * - SparseCholesky 10 | * - IterativeLinearSolvers 11 | * 12 | * \code 13 | * #include 14 | * \endcode 15 | */ 16 | 17 | #include "SparseCore" 18 | #include "OrderingMethods" 19 | #include "SparseCholesky" 20 | #include "IterativeLinearSolvers" 21 | 22 | #endif // EIGEN_SPARSE_MODULE_H 23 | 24 | -------------------------------------------------------------------------------- /docker/docker-jupyter/jupyter_notebook_config.py: -------------------------------------------------------------------------------- 1 | import os 2 | from IPython.lib import passwd 3 | 4 | #c = c # pylint:disable=undefined-variable 5 | c = get_config() 6 | c.NotebookApp.ip = '0.0.0.0' 7 | c.NotebookApp.port = int(os.getenv('PORT', 8888)) 8 | c.NotebookApp.open_browser = False 9 | 10 | # sets a password if PASSWORD is set in the environment 11 | if 'PASSWORD' in os.environ: 12 | password = os.environ['PASSWORD'] 13 | if password: 14 | c.NotebookApp.password = passwd(password) 15 | else: 16 | c.NotebookApp.password = '' 17 | c.NotebookApp.token = '' 18 | del os.environ['PASSWORD'] 19 | -------------------------------------------------------------------------------- /configs/quick_schedules/rpn_R_50_C4_quick.yaml: -------------------------------------------------------------------------------- 1 | MODEL: 2 | META_ARCHITECTURE: "GeneralizedRCNN" 3 | WEIGHT: "catalog://ImageNetPretrained/MSRA/R-50" 4 | RPN_ONLY: True 5 | RPN: 6 | PRE_NMS_TOP_N_TEST: 12000 7 | POST_NMS_TOP_N_TEST: 2000 8 | DATASETS: 9 | TRAIN: ("coco_2014_minival",) 10 | TEST: ("coco_2014_minival",) 11 | INPUT: 12 | MIN_SIZE_TRAIN: (600,) 13 | MAX_SIZE_TRAIN: 1000 14 | MIN_SIZE_TEST: 800 15 | MAX_SIZE_TEST: 1000 16 | SOLVER: 17 | BASE_LR: 0.005 18 | WEIGHT_DECAY: 0.0001 19 | STEPS: (1500,) 20 | MAX_ITER: 2000 21 | IMS_PER_BATCH: 4 22 | TEST: 23 | IMS_PER_BATCH: 2 24 | -------------------------------------------------------------------------------- /maskrcnn_benchmark/data/datasets/evaluation/voc/__init__.py: -------------------------------------------------------------------------------- 1 | import logging 2 | 3 | from .voc_eval import do_voc_evaluation 4 | 5 | 6 | def voc_evaluation(dataset, predictions, output_folder, box_only, **_): 7 | logger = logging.getLogger("maskrcnn_benchmark.inference") 8 | if box_only: 9 | logger.warning("voc evaluation doesn't support box_only, ignored.") 10 | logger.info("performing voc evaluation, ignored iou_types.") 11 | return do_voc_evaluation( 12 | dataset=dataset, 13 | predictions=predictions, 14 | output_folder=output_folder, 15 | logger=logger, 16 | ) 17 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/decompress.m: -------------------------------------------------------------------------------- 1 | % Wrapper for probImage decompress 2 | function decompressed = decompress(file_path) 3 | 4 | addpath('auxilary'); 5 | 6 | if ~(exist(file_path)) 7 | error('Unary file given does not exist') 8 | end 9 | 10 | cpp_file = 'decompress_mex.cpp'; 11 | out_file = 'decompress_mex'; 12 | 13 | extra_arguments = {}; 14 | 15 | % Additional files to be compiled. 16 | dec_dir = ['probImage' filesep]; 17 | sources = {[dec_dir 'probimage.cpp']}; 18 | 19 | % Only compile if files have changed 20 | compile(cpp_file, out_file, sources, extra_arguments) 21 | 22 | decompressed = decompress_mex(file_path); 23 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/OrderingMethods: -------------------------------------------------------------------------------- 1 | #ifndef EIGEN_ORDERINGMETHODS_MODULE_H 2 | #define EIGEN_ORDERINGMETHODS_MODULE_H 3 | 4 | #include "SparseCore" 5 | 6 | #include "src/Core/util/DisableStupidWarnings.h" 7 | 8 | /** \ingroup Sparse_modules 9 | * \defgroup OrderingMethods_Module OrderingMethods module 10 | * 11 | * This module is currently for internal use only. 12 | * 13 | * 14 | * \code 15 | * #include 16 | * \endcode 17 | */ 18 | 19 | #include "src/OrderingMethods/Amd.h" 20 | 21 | #include "src/Core/util/ReenableStupidWarnings.h" 22 | 23 | #endif // EIGEN_ORDERINGMETHODS_MODULE_H 24 | -------------------------------------------------------------------------------- /configs/quick_schedules/e2e_faster_rcnn_R_50_C4_quick.yaml: -------------------------------------------------------------------------------- 1 | MODEL: 2 | META_ARCHITECTURE: "GeneralizedRCNN" 3 | WEIGHT: "catalog://ImageNetPretrained/MSRA/R-50" 4 | RPN: 5 | PRE_NMS_TOP_N_TEST: 6000 6 | POST_NMS_TOP_N_TEST: 1000 7 | ROI_HEADS: 8 | BATCH_SIZE_PER_IMAGE: 256 9 | DATASETS: 10 | TRAIN: ("coco_2014_minival",) 11 | TEST: ("coco_2014_minival",) 12 | INPUT: 13 | MIN_SIZE_TRAIN: (600,) 14 | MAX_SIZE_TRAIN: 1000 15 | MIN_SIZE_TEST: 800 16 | MAX_SIZE_TEST: 1000 17 | SOLVER: 18 | BASE_LR: 0.005 19 | WEIGHT_DECAY: 0.0001 20 | STEPS: (1500,) 21 | MAX_ITER: 2000 22 | IMS_PER_BATCH: 2 23 | TEST: 24 | IMS_PER_BATCH: 2 25 | -------------------------------------------------------------------------------- /maskrcnn_benchmark/csrc/cpu/vision.h: -------------------------------------------------------------------------------- 1 | // Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. 2 | #pragma once 3 | #include 4 | 5 | 6 | at::Tensor ROIAlign_forward_cpu(const at::Tensor& input, 7 | const at::Tensor& rois, 8 | const float spatial_scale, 9 | const int pooled_height, 10 | const int pooled_width, 11 | const int sampling_ratio); 12 | 13 | 14 | at::Tensor nms_cpu(const at::Tensor& dets, 15 | const at::Tensor& scores, 16 | const float threshold); 17 | -------------------------------------------------------------------------------- /configs/rpn_R_50_FPN_1x.yaml: -------------------------------------------------------------------------------- 1 | MODEL: 2 | META_ARCHITECTURE: "GeneralizedRCNN" 3 | WEIGHT: "catalog://ImageNetPretrained/MSRA/R-50" 4 | RPN_ONLY: True 5 | BACKBONE: 6 | CONV_BODY: "R-50-FPN" 7 | RESNETS: 8 | BACKBONE_OUT_CHANNELS: 256 9 | RPN: 10 | USE_FPN: True 11 | ANCHOR_STRIDE: (4, 8, 16, 32, 64) 12 | PRE_NMS_TOP_N_TEST: 1000 13 | POST_NMS_TOP_N_TEST: 2000 14 | FPN_POST_NMS_TOP_N_TEST: 2000 15 | DATASETS: 16 | TRAIN: ("coco_2014_train", "coco_2014_valminusminival") 17 | TEST: ("coco_2014_minival",) 18 | DATALOADER: 19 | SIZE_DIVISIBILITY: 32 20 | SOLVER: 21 | BASE_LR: 0.02 22 | WEIGHT_DECAY: 0.0001 23 | STEPS: (60000, 80000) 24 | MAX_ITER: 90000 25 | -------------------------------------------------------------------------------- /configs/rpn_R_101_FPN_1x.yaml: -------------------------------------------------------------------------------- 1 | MODEL: 2 | META_ARCHITECTURE: "GeneralizedRCNN" 3 | WEIGHT: "catalog://ImageNetPretrained/MSRA/R-101" 4 | RPN_ONLY: True 5 | BACKBONE: 6 | CONV_BODY: "R-101-FPN" 7 | RESNETS: 8 | BACKBONE_OUT_CHANNELS: 256 9 | RPN: 10 | USE_FPN: True 11 | ANCHOR_STRIDE: (4, 8, 16, 32, 64) 12 | PRE_NMS_TOP_N_TEST: 1000 13 | POST_NMS_TOP_N_TEST: 2000 14 | FPN_POST_NMS_TOP_N_TEST: 2000 15 | DATASETS: 16 | TRAIN: ("coco_2014_train", "coco_2014_valminusminival") 17 | TEST: ("coco_2014_minival",) 18 | DATALOADER: 19 | SIZE_DIVISIBILITY: 32 20 | SOLVER: 21 | BASE_LR: 0.02 22 | WEIGHT_DECAY: 0.0001 23 | STEPS: (60000, 80000) 24 | MAX_ITER: 90000 25 | -------------------------------------------------------------------------------- /maskrcnn_benchmark/data/datasets/evaluation/coco/__init__.py: -------------------------------------------------------------------------------- 1 | from .coco_eval import do_coco_evaluation 2 | 3 | 4 | def coco_evaluation( 5 | dataset, 6 | predictions, 7 | output_folder, 8 | box_only, 9 | iou_types, 10 | expected_results, 11 | expected_results_sigma_tol, 12 | save_pred_as_mat, 13 | ): 14 | return do_coco_evaluation( 15 | dataset=dataset, 16 | predictions=predictions, 17 | box_only=box_only, 18 | output_folder=output_folder, 19 | iou_types=iou_types, 20 | expected_results=expected_results, 21 | expected_results_sigma_tol=expected_results_sigma_tol, 22 | save_pred_as_mat=save_pred_as_mat, 23 | ) 24 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/examples/CMakeLists.txt: -------------------------------------------------------------------------------- 1 | add_executable( dense_inference ppm.cpp common.cpp dense_inference.cpp ) 2 | target_link_libraries( dense_inference densecrf ) 3 | 4 | add_executable( dense_learning ppm.cpp common.cpp dense_learning.cpp ) 5 | target_link_libraries( dense_learning densecrf optimization ) 6 | 7 | add_executable( test_optimization test_optimization.cpp ) 8 | target_link_libraries( test_optimization optimization ) 9 | 10 | add_executable( test_permutohedral test_permutohedral.cpp ) 11 | target_link_libraries( test_permutohedral densecrf optimization ) 12 | 13 | add_executable( test_pairwise test_pairwise.cpp ) 14 | target_link_libraries( test_pairwise densecrf optimization ) 15 | -------------------------------------------------------------------------------- /tests/test_configs.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. 2 | 3 | import unittest 4 | import glob 5 | import os 6 | import utils 7 | 8 | 9 | class TestConfigs(unittest.TestCase): 10 | def test_configs_load(self): 11 | ''' Make sure configs are loadable ''' 12 | 13 | cfg_root_path = utils.get_config_root_path() 14 | files = glob.glob( 15 | os.path.join(cfg_root_path, "./**/*.yaml"), recursive=True) 16 | self.assertGreater(len(files), 0) 17 | 18 | for fn in files: 19 | print('Loading {}...'.format(fn)) 20 | utils.load_config_from_file(fn) 21 | 22 | 23 | if __name__ == "__main__": 24 | unittest.main() 25 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/Householder: -------------------------------------------------------------------------------- 1 | #ifndef EIGEN_HOUSEHOLDER_MODULE_H 2 | #define EIGEN_HOUSEHOLDER_MODULE_H 3 | 4 | #include "Core" 5 | 6 | #include "src/Core/util/DisableStupidWarnings.h" 7 | 8 | /** \defgroup Householder_Module Householder module 9 | * This module provides Householder transformations. 10 | * 11 | * \code 12 | * #include 13 | * \endcode 14 | */ 15 | 16 | #include "src/Householder/Householder.h" 17 | #include "src/Householder/HouseholderSequence.h" 18 | #include "src/Householder/BlockHouseholder.h" 19 | 20 | #include "src/Core/util/ReenableStupidWarnings.h" 21 | 22 | #endif // EIGEN_HOUSEHOLDER_MODULE_H 23 | /* vim: set filetype=cpp et sw=2 ts=2 ai: */ 24 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/CMakeLists.txt: -------------------------------------------------------------------------------- 1 | include(RegexUtils) 2 | test_escape_string_as_regex() 3 | 4 | file(GLOB Eigen_directory_files "*") 5 | 6 | escape_string_as_regex(ESCAPED_CMAKE_CURRENT_SOURCE_DIR "${CMAKE_CURRENT_SOURCE_DIR}") 7 | 8 | foreach(f ${Eigen_directory_files}) 9 | if(NOT f MATCHES "\\.txt" AND NOT f MATCHES "${ESCAPED_CMAKE_CURRENT_SOURCE_DIR}/[.].+" AND NOT f MATCHES "${ESCAPED_CMAKE_CURRENT_SOURCE_DIR}/src") 10 | list(APPEND Eigen_directory_files_to_install ${f}) 11 | endif() 12 | endforeach(f ${Eigen_directory_files}) 13 | 14 | install(FILES 15 | ${Eigen_directory_files_to_install} 16 | DESTINATION ${INCLUDE_INSTALL_DIR}/Eigen COMPONENT Devel 17 | ) 18 | 19 | add_subdirectory(src) 20 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/src/Eigen2Support/Macros.h: -------------------------------------------------------------------------------- 1 | // This file is part of Eigen, a lightweight C++ template library 2 | // for linear algebra. 3 | // 4 | // Copyright (C) 2011 Benoit Jacob 5 | // 6 | // This Source Code Form is subject to the terms of the Mozilla 7 | // Public License v. 2.0. If a copy of the MPL was not distributed 8 | // with this file, You can obtain one at http://mozilla.org/MPL/2.0/. 9 | 10 | #ifndef EIGEN2_MACROS_H 11 | #define EIGEN2_MACROS_H 12 | 13 | #define ei_assert eigen_assert 14 | #define ei_internal_assert eigen_internal_assert 15 | 16 | #define EIGEN_ALIGN_128 EIGEN_ALIGN16 17 | 18 | #define EIGEN_ARCH_WANTS_ALIGNMENT EIGEN_ALIGN_STATICALLY 19 | 20 | #endif // EIGEN2_MACROS_H 21 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/external/liblbfgs/autogen.sh: -------------------------------------------------------------------------------- 1 | #!/bin/sh 2 | # $Id$ 3 | 4 | if [ "$1" = "--force" ]; 5 | then 6 | FORCE=--force 7 | NOFORCE= 8 | FORCE_MISSING=--force-missing 9 | else 10 | FORCE= 11 | NOFORCE=--no-force 12 | FORCE_MISSING= 13 | fi 14 | 15 | libtoolize --copy $FORCE 2>&1 | sed '/^You should/d' || { 16 | echo "libtoolize failed!" 17 | exit 1 18 | } 19 | 20 | aclocal $FORCE || { 21 | echo "aclocal failed!" 22 | exit 1 23 | } 24 | 25 | autoheader $FORCE || { 26 | echo "autoheader failed!" 27 | exit 1 28 | } 29 | 30 | automake -a -c $NOFORCE || { 31 | echo "automake failed!" 32 | exit 1 33 | } 34 | 35 | autoconf $FORCE || { 36 | echo "autoconf failed!" 37 | exit 1 38 | } 39 | -------------------------------------------------------------------------------- /maskrcnn_benchmark/utils/cv2_util.py: -------------------------------------------------------------------------------- 1 | """ 2 | Module for cv2 utility functions and maintaining version compatibility 3 | between 3.x and 4.x 4 | """ 5 | import cv2 6 | 7 | 8 | def findContours(*args, **kwargs): 9 | """ 10 | Wraps cv2.findContours to maintain compatiblity between versions 11 | 3 and 4 12 | 13 | Returns: 14 | contours, hierarchy 15 | """ 16 | if cv2.__version__.startswith('4'): 17 | contours, hierarchy = cv2.findContours(*args, **kwargs) 18 | elif cv2.__version__.startswith('3'): 19 | _, contours, hierarchy = cv2.findContours(*args, **kwargs) 20 | else: 21 | raise AssertionError( 22 | 'cv2 must be either version 3 or 4 to call this method') 23 | 24 | return contours, hierarchy 25 | -------------------------------------------------------------------------------- /configs/quick_schedules/e2e_mask_rcnn_R_50_C4_quick.yaml: -------------------------------------------------------------------------------- 1 | MODEL: 2 | META_ARCHITECTURE: "GeneralizedRCNN" 3 | WEIGHT: "catalog://ImageNetPretrained/MSRA/R-50" 4 | RPN: 5 | PRE_NMS_TOP_N_TEST: 6000 6 | POST_NMS_TOP_N_TEST: 1000 7 | ROI_HEADS: 8 | BATCH_SIZE_PER_IMAGE: 256 9 | ROI_MASK_HEAD: 10 | PREDICTOR: "MaskRCNNC4Predictor" 11 | SHARE_BOX_FEATURE_EXTRACTOR: True 12 | MASK_ON: True 13 | DATASETS: 14 | TRAIN: ("coco_2014_minival",) 15 | TEST: ("coco_2014_minival",) 16 | INPUT: 17 | MIN_SIZE_TRAIN: (600,) 18 | MAX_SIZE_TRAIN: 1000 19 | MIN_SIZE_TEST: 800 20 | MAX_SIZE_TEST: 1000 21 | SOLVER: 22 | BASE_LR: 0.005 23 | WEIGHT_DECAY: 0.0001 24 | STEPS: (1500,) 25 | MAX_ITER: 2000 26 | IMS_PER_BATCH: 4 27 | TEST: 28 | IMS_PER_BATCH: 2 29 | -------------------------------------------------------------------------------- /maskrcnn_benchmark/data/collate_batch.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. 2 | from maskrcnn_benchmark.structures.image_list import to_image_list 3 | 4 | 5 | class BatchCollator(object): 6 | """ 7 | From a list of samples from the dataset, 8 | returns the batched images and targets. 9 | This should be passed to the DataLoader 10 | """ 11 | 12 | def __init__(self, size_divisible=0): 13 | self.size_divisible = size_divisible 14 | 15 | def __call__(self, batch): 16 | transposed_batch = list(zip(*batch)) 17 | images = to_image_list(transposed_batch[0], self.size_divisible) 18 | targets = transposed_batch[1] 19 | img_ids = transposed_batch[2] 20 | return images, targets, img_ids 21 | -------------------------------------------------------------------------------- /configs/rpn_X_101_32x8d_FPN_1x.yaml: -------------------------------------------------------------------------------- 1 | MODEL: 2 | META_ARCHITECTURE: "GeneralizedRCNN" 3 | WEIGHT: "catalog://ImageNetPretrained/FAIR/20171220/X-101-32x8d" 4 | RPN_ONLY: True 5 | BACKBONE: 6 | CONV_BODY: "R-101-FPN" 7 | RESNETS: 8 | BACKBONE_OUT_CHANNELS: 256 9 | STRIDE_IN_1X1: False 10 | NUM_GROUPS: 32 11 | WIDTH_PER_GROUP: 8 12 | RPN: 13 | USE_FPN: True 14 | ANCHOR_STRIDE: (4, 8, 16, 32, 64) 15 | PRE_NMS_TOP_N_TEST: 1000 16 | POST_NMS_TOP_N_TEST: 2000 17 | FPN_POST_NMS_TOP_N_TEST: 2000 18 | DATASETS: 19 | TRAIN: ("coco_2014_train", "coco_2014_valminusminival") 20 | TEST: ("coco_2014_minival",) 21 | DATALOADER: 22 | SIZE_DIVISIBILITY: 32 23 | SOLVER: 24 | BASE_LR: 0.02 25 | WEIGHT_DECAY: 0.0001 26 | STEPS: (60000, 80000) 27 | MAX_ITER: 90000 28 | -------------------------------------------------------------------------------- /maskrcnn_benchmark/csrc/vision.cpp: -------------------------------------------------------------------------------- 1 | // Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. 2 | #include "nms.h" 3 | #include "ROIAlign.h" 4 | #include "ROIPool.h" 5 | #include "SigmoidFocalLoss.h" 6 | 7 | PYBIND11_MODULE(TORCH_EXTENSION_NAME, m) { 8 | m.def("nms", &nms, "non-maximum suppression"); 9 | m.def("roi_align_forward", &ROIAlign_forward, "ROIAlign_forward"); 10 | m.def("roi_align_backward", &ROIAlign_backward, "ROIAlign_backward"); 11 | m.def("roi_pool_forward", &ROIPool_forward, "ROIPool_forward"); 12 | m.def("roi_pool_backward", &ROIPool_backward, "ROIPool_backward"); 13 | m.def("sigmoid_focalloss_forward", &SigmoidFocalLoss_forward, "SigmoidFocalLoss_forward"); 14 | m.def("sigmoid_focalloss_backward", &SigmoidFocalLoss_backward, "SigmoidFocalLoss_backward"); 15 | } 16 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/Jacobi: -------------------------------------------------------------------------------- 1 | #ifndef EIGEN_JACOBI_MODULE_H 2 | #define EIGEN_JACOBI_MODULE_H 3 | 4 | #include "Core" 5 | 6 | #include "src/Core/util/DisableStupidWarnings.h" 7 | 8 | /** \defgroup Jacobi_Module Jacobi module 9 | * This module provides Jacobi and Givens rotations. 10 | * 11 | * \code 12 | * #include 13 | * \endcode 14 | * 15 | * In addition to listed classes, it defines the two following MatrixBase methods to apply a Jacobi or Givens rotation: 16 | * - MatrixBase::applyOnTheLeft() 17 | * - MatrixBase::applyOnTheRight(). 18 | */ 19 | 20 | #include "src/Jacobi/Jacobi.h" 21 | 22 | #include "src/Core/util/ReenableStupidWarnings.h" 23 | 24 | #endif // EIGEN_JACOBI_MODULE_H 25 | /* vim: set filetype=cpp et sw=2 ts=2 ai: */ 26 | 27 | -------------------------------------------------------------------------------- /maskrcnn_benchmark/layers/__init__.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. 2 | import torch 3 | 4 | from .batch_norm import FrozenBatchNorm2d 5 | from .misc import Conv2d 6 | from .misc import ConvTranspose2d 7 | from .misc import BatchNorm2d 8 | from .misc import interpolate 9 | from .nms import nms 10 | from .roi_align import ROIAlign 11 | from .roi_align import roi_align 12 | from .roi_pool import ROIPool 13 | from .roi_pool import roi_pool 14 | from .smooth_l1_loss import smooth_l1_loss 15 | from .sigmoid_focal_loss import SigmoidFocalLoss 16 | 17 | __all__ = ["nms", "roi_align", "ROIAlign", "roi_pool", "ROIPool", 18 | "smooth_l1_loss", "Conv2d", "ConvTranspose2d", "interpolate", 19 | "BatchNorm2d", "FrozenBatchNorm2d", "SigmoidFocalLoss" 20 | ] 21 | 22 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/StdList: -------------------------------------------------------------------------------- 1 | // This file is part of Eigen, a lightweight C++ template library 2 | // for linear algebra. 3 | // 4 | // Copyright (C) 2009 Hauke Heibel 5 | // 6 | // This Source Code Form is subject to the terms of the Mozilla 7 | // Public License v. 2.0. If a copy of the MPL was not distributed 8 | // with this file, You can obtain one at http://mozilla.org/MPL/2.0/. 9 | 10 | #ifndef EIGEN_STDLIST_MODULE_H 11 | #define EIGEN_STDLIST_MODULE_H 12 | 13 | #include "Core" 14 | #include 15 | 16 | #if (defined(_MSC_VER) && defined(_WIN64)) /* MSVC auto aligns in 64 bit builds */ 17 | 18 | #define EIGEN_DEFINE_STL_LIST_SPECIALIZATION(...) 19 | 20 | #else 21 | 22 | #include "src/StlSupport/StdList.h" 23 | 24 | #endif 25 | 26 | #endif // EIGEN_STDLIST_MODULE_H 27 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/QtAlignedMalloc: -------------------------------------------------------------------------------- 1 | 2 | #ifndef EIGEN_QTMALLOC_MODULE_H 3 | #define EIGEN_QTMALLOC_MODULE_H 4 | 5 | #include "Core" 6 | 7 | #if (!EIGEN_MALLOC_ALREADY_ALIGNED) 8 | 9 | #include "src/Core/util/DisableStupidWarnings.h" 10 | 11 | void *qMalloc(size_t size) 12 | { 13 | return Eigen::internal::aligned_malloc(size); 14 | } 15 | 16 | void qFree(void *ptr) 17 | { 18 | Eigen::internal::aligned_free(ptr); 19 | } 20 | 21 | void *qRealloc(void *ptr, size_t size) 22 | { 23 | void* newPtr = Eigen::internal::aligned_malloc(size); 24 | memcpy(newPtr, ptr, size); 25 | Eigen::internal::aligned_free(ptr); 26 | return newPtr; 27 | } 28 | 29 | #include "src/Core/util/ReenableStupidWarnings.h" 30 | 31 | #endif 32 | 33 | #endif // EIGEN_QTMALLOC_MODULE_H 34 | /* vim: set filetype=cpp et sw=2 ts=2 ai: */ 35 | -------------------------------------------------------------------------------- /configs/caffe2/e2e_faster_rcnn_R_101_FPN_1x_caffe2.yaml: -------------------------------------------------------------------------------- 1 | MODEL: 2 | META_ARCHITECTURE: "GeneralizedRCNN" 3 | WEIGHT: "catalog://Caffe2Detectron/COCO/35857890/e2e_faster_rcnn_R-101-FPN_1x" 4 | BACKBONE: 5 | CONV_BODY: "R-101-FPN" 6 | RESNETS: 7 | BACKBONE_OUT_CHANNELS: 256 8 | RPN: 9 | USE_FPN: True 10 | ANCHOR_STRIDE: (4, 8, 16, 32, 64) 11 | PRE_NMS_TOP_N_TRAIN: 2000 12 | PRE_NMS_TOP_N_TEST: 1000 13 | POST_NMS_TOP_N_TEST: 1000 14 | FPN_POST_NMS_TOP_N_TEST: 1000 15 | ROI_HEADS: 16 | USE_FPN: True 17 | ROI_BOX_HEAD: 18 | POOLER_RESOLUTION: 7 19 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 20 | POOLER_SAMPLING_RATIO: 2 21 | FEATURE_EXTRACTOR: "FPN2MLPFeatureExtractor" 22 | PREDICTOR: "FPNPredictor" 23 | DATASETS: 24 | TEST: ("coco_2014_minival",) 25 | DATALOADER: 26 | SIZE_DIVISIBILITY: 32 27 | -------------------------------------------------------------------------------- /configs/caffe2/e2e_faster_rcnn_R_50_FPN_1x_caffe2.yaml: -------------------------------------------------------------------------------- 1 | MODEL: 2 | META_ARCHITECTURE: "GeneralizedRCNN" 3 | WEIGHT: "catalog://Caffe2Detectron/COCO/35857345/e2e_faster_rcnn_R-50-FPN_1x" 4 | BACKBONE: 5 | CONV_BODY: "R-50-FPN" 6 | RESNETS: 7 | BACKBONE_OUT_CHANNELS: 256 8 | RPN: 9 | USE_FPN: True 10 | ANCHOR_STRIDE: (4, 8, 16, 32, 64) 11 | PRE_NMS_TOP_N_TRAIN: 2000 12 | PRE_NMS_TOP_N_TEST: 1000 13 | POST_NMS_TOP_N_TEST: 1000 14 | FPN_POST_NMS_TOP_N_TEST: 1000 15 | ROI_HEADS: 16 | USE_FPN: True 17 | ROI_BOX_HEAD: 18 | POOLER_RESOLUTION: 7 19 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 20 | POOLER_SAMPLING_RATIO: 2 21 | FEATURE_EXTRACTOR: "FPN2MLPFeatureExtractor" 22 | PREDICTOR: "FPNPredictor" 23 | DATASETS: 24 | TEST: ("coco_2014_minival",) 25 | DATALOADER: 26 | SIZE_DIVISIBILITY: 32 27 | -------------------------------------------------------------------------------- /maskrcnn_benchmark/csrc/nms.h: -------------------------------------------------------------------------------- 1 | // Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. 2 | #pragma once 3 | #include "cpu/vision.h" 4 | 5 | #ifdef WITH_CUDA 6 | #include "cuda/vision.h" 7 | #endif 8 | 9 | 10 | at::Tensor nms(const at::Tensor& dets, 11 | const at::Tensor& scores, 12 | const float threshold) { 13 | 14 | if (dets.type().is_cuda()) { 15 | #ifdef WITH_CUDA 16 | // TODO raise error if not compiled with CUDA 17 | if (dets.numel() == 0) 18 | return at::empty({0}, dets.options().dtype(at::kLong).device(at::kCPU)); 19 | auto b = at::cat({dets, scores.unsqueeze(1)}, 1); 20 | return nms_cuda(b, threshold); 21 | #else 22 | AT_ERROR("Not compiled with GPU support"); 23 | #endif 24 | } 25 | 26 | at::Tensor result = nms_cpu(dets, scores, threshold); 27 | return result; 28 | } 29 | -------------------------------------------------------------------------------- /configs/quick_schedules/rpn_R_50_FPN_quick.yaml: -------------------------------------------------------------------------------- 1 | MODEL: 2 | META_ARCHITECTURE: "GeneralizedRCNN" 3 | WEIGHT: "catalog://ImageNetPretrained/MSRA/R-50" 4 | RPN_ONLY: True 5 | BACKBONE: 6 | CONV_BODY: "R-50-FPN" 7 | RESNETS: 8 | BACKBONE_OUT_CHANNELS: 256 9 | RPN: 10 | USE_FPN: True 11 | ANCHOR_STRIDE: (4, 8, 16, 32, 64) 12 | PRE_NMS_TOP_N_TEST: 1000 13 | POST_NMS_TOP_N_TEST: 2000 14 | FPN_POST_NMS_TOP_N_TEST: 2000 15 | DATASETS: 16 | TRAIN: ("coco_2014_minival",) 17 | TEST: ("coco_2014_minival",) 18 | INPUT: 19 | MIN_SIZE_TRAIN: (600,) 20 | MAX_SIZE_TRAIN: 1000 21 | MIN_SIZE_TEST: 800 22 | MAX_SIZE_TEST: 1000 23 | DATALOADER: 24 | SIZE_DIVISIBILITY: 32 25 | SOLVER: 26 | BASE_LR: 0.005 27 | WEIGHT_DECAY: 0.0001 28 | STEPS: (1500,) 29 | MAX_ITER: 2000 30 | IMS_PER_BATCH: 4 31 | TEST: 32 | IMS_PER_BATCH: 2 33 | -------------------------------------------------------------------------------- /configs/gn_baselines/README.md: -------------------------------------------------------------------------------- 1 | ### Group Normalization 2 | 1 [Group Normalization](https://arxiv.org/abs/1803.08494) 3 | 2 [Rethinking ImageNet Pre-training](https://arxiv.org/abs/1811.08883) 4 | 3 [official code](https://github.com/facebookresearch/Detectron/blob/master/projects/GN/README.md) 5 | 6 | 7 | ### Performance 8 | | case | Type | lr schd | im/gpu | bbox AP | mask AP | 9 | |----------------------------|:------------:|:---------:|:-------:|:-------:|:-------:| 10 | | R-50-FPN, GN (paper) | finetune | 2x | 2 | 40.3 | 35.7 | 11 | | R-50-FPN, GN (implement) | finetune | 2x | 2 | 40.2 | 36.0 | 12 | | R-50-FPN, GN (paper) | from scratch | 3x | 2 | 39.5 | 35.2 | 13 | | R-50-FPN, GN (implement) | from scratch | 3x | 2 | 38.9 | 35.1 | 14 | -------------------------------------------------------------------------------- /matlab/DenseCRF/examples/example.m: -------------------------------------------------------------------------------- 1 | clear all; close all; 2 | addpath('..'); 3 | 4 | image = imread('../include/densecrf/examples/im2.ppm'); 5 | unary = -single(image); 6 | D = Densecrf(); 7 | D.SetImage(image); 8 | D.SetUnary(unary); 9 | % Some settings. 10 | D.gaussian_x_stddev = 3; 11 | D.gaussian_y_stddev = 3; 12 | D.gaussian_weight = 1; 13 | 14 | D.bilateral_x_stddev = 60; 15 | D.bilateral_y_stddev = 60; 16 | D.bilateral_r_stddev = 10; 17 | D.bilateral_g_stddev = 10; 18 | D.bilateral_b_stddev = 10; 19 | D.bilateral_weight = 1; 20 | 21 | 22 | %% Threhold 23 | % figure(1); 24 | % D.threshold; 25 | % D.display(); 26 | 27 | %% Meanfield 28 | figure(2); 29 | D.mean_field; 30 | D.display(); 31 | 32 | % %% TRW-S (adding very few edges) 33 | % figure() 34 | % 35 | % % set to zero to solve the densecrf problem (requires alot of memory) 36 | % D.min_pairwise_cost = 1; 37 | % D.trws; 38 | % D.display() -------------------------------------------------------------------------------- /maskrcnn_benchmark/data/datasets/concat_dataset.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. 2 | import bisect 3 | 4 | from torch.utils.data.dataset import ConcatDataset as _ConcatDataset 5 | 6 | 7 | class ConcatDataset(_ConcatDataset): 8 | """ 9 | Same as torch.utils.data.dataset.ConcatDataset, but exposes an extra 10 | method for querying the sizes of the image 11 | """ 12 | 13 | def get_idxs(self, idx): 14 | dataset_idx = bisect.bisect_right(self.cumulative_sizes, idx) 15 | if dataset_idx == 0: 16 | sample_idx = idx 17 | else: 18 | sample_idx = idx - self.cumulative_sizes[dataset_idx - 1] 19 | return dataset_idx, sample_idx 20 | 21 | def get_img_info(self, idx): 22 | dataset_idx, sample_idx = self.get_idxs(idx) 23 | return self.datasets[dataset_idx].get_img_info(sample_idx) 24 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/LeastSquares: -------------------------------------------------------------------------------- 1 | #ifndef EIGEN_REGRESSION_MODULE_H 2 | #define EIGEN_REGRESSION_MODULE_H 3 | 4 | #ifndef EIGEN2_SUPPORT 5 | #error LeastSquares is only available in Eigen2 support mode (define EIGEN2_SUPPORT) 6 | #endif 7 | 8 | // exclude from normal eigen3-only documentation 9 | #ifdef EIGEN2_SUPPORT 10 | 11 | #include "Core" 12 | 13 | #include "src/Core/util/DisableStupidWarnings.h" 14 | 15 | #include "Eigenvalues" 16 | #include "Geometry" 17 | 18 | /** \defgroup LeastSquares_Module LeastSquares module 19 | * This module provides linear regression and related features. 20 | * 21 | * \code 22 | * #include 23 | * \endcode 24 | */ 25 | 26 | #include "src/Eigen2Support/LeastSquares.h" 27 | 28 | #include "src/Core/util/ReenableStupidWarnings.h" 29 | 30 | #endif // EIGEN2_SUPPORT 31 | 32 | #endif // EIGEN_REGRESSION_MODULE_H 33 | -------------------------------------------------------------------------------- /maskrcnn_benchmark/utils/logger.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. 2 | import logging 3 | import os 4 | import sys 5 | 6 | 7 | def setup_logger(name, save_dir, distributed_rank, filename="log.txt"): 8 | logger = logging.getLogger(name) 9 | logger.setLevel(logging.DEBUG) 10 | # don't log results for the non-master process 11 | if distributed_rank > 0: 12 | return logger 13 | ch = logging.StreamHandler(stream=sys.stdout) 14 | ch.setLevel(logging.DEBUG) 15 | formatter = logging.Formatter("%(asctime)s %(name)s %(levelname)s: %(message)s") 16 | ch.setFormatter(formatter) 17 | logger.addHandler(ch) 18 | 19 | if save_dir: 20 | fh = logging.FileHandler(os.path.join(save_dir, filename)) 21 | fh.setLevel(logging.DEBUG) 22 | fh.setFormatter(formatter) 23 | logger.addHandler(fh) 24 | 25 | return logger 26 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/StdDeque: -------------------------------------------------------------------------------- 1 | // This file is part of Eigen, a lightweight C++ template library 2 | // for linear algebra. 3 | // 4 | // Copyright (C) 2009 Gael Guennebaud 5 | // Copyright (C) 2009 Hauke Heibel 6 | // 7 | // This Source Code Form is subject to the terms of the Mozilla 8 | // Public License v. 2.0. If a copy of the MPL was not distributed 9 | // with this file, You can obtain one at http://mozilla.org/MPL/2.0/. 10 | 11 | #ifndef EIGEN_STDDEQUE_MODULE_H 12 | #define EIGEN_STDDEQUE_MODULE_H 13 | 14 | #include "Core" 15 | #include 16 | 17 | #if (defined(_MSC_VER) && defined(_WIN64)) /* MSVC auto aligns in 64 bit builds */ 18 | 19 | #define EIGEN_DEFINE_STL_DEQUE_SPECIALIZATION(...) 20 | 21 | #else 22 | 23 | #include "src/StlSupport/StdDeque.h" 24 | 25 | #endif 26 | 27 | #endif // EIGEN_STDDEQUE_MODULE_H 28 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/StdVector: -------------------------------------------------------------------------------- 1 | // This file is part of Eigen, a lightweight C++ template library 2 | // for linear algebra. 3 | // 4 | // Copyright (C) 2009 Gael Guennebaud 5 | // Copyright (C) 2009 Hauke Heibel 6 | // 7 | // This Source Code Form is subject to the terms of the Mozilla 8 | // Public License v. 2.0. If a copy of the MPL was not distributed 9 | // with this file, You can obtain one at http://mozilla.org/MPL/2.0/. 10 | 11 | #ifndef EIGEN_STDVECTOR_MODULE_H 12 | #define EIGEN_STDVECTOR_MODULE_H 13 | 14 | #include "Core" 15 | #include 16 | 17 | #if (defined(_MSC_VER) && defined(_WIN64)) /* MSVC auto aligns in 64 bit builds */ 18 | 19 | #define EIGEN_DEFINE_STL_VECTOR_SPECIALIZATION(...) 20 | 21 | #else 22 | 23 | #include "src/StlSupport/StdVector.h" 24 | 25 | #endif 26 | 27 | #endif // EIGEN_STDVECTOR_MODULE_H 28 | -------------------------------------------------------------------------------- /tests/test_metric_logger.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. 2 | import unittest 3 | 4 | from maskrcnn_benchmark.utils.metric_logger import MetricLogger 5 | 6 | 7 | class TestMetricLogger(unittest.TestCase): 8 | def test_update(self): 9 | meter = MetricLogger() 10 | for i in range(10): 11 | meter.update(metric=float(i)) 12 | 13 | m = meter.meters["metric"] 14 | self.assertEqual(m.count, 10) 15 | self.assertEqual(m.total, 45) 16 | self.assertEqual(m.median, 4) 17 | self.assertEqual(m.avg, 4.5) 18 | 19 | def test_no_attr(self): 20 | meter = MetricLogger() 21 | _ = meter.meters 22 | _ = meter.delimiter 23 | def broken(): 24 | _ = meter.not_existent 25 | self.assertRaises(AttributeError, broken) 26 | 27 | if __name__ == "__main__": 28 | unittest.main() 29 | -------------------------------------------------------------------------------- /maskrcnn_benchmark/layers/batch_norm.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. 2 | import torch 3 | from torch import nn 4 | 5 | 6 | class FrozenBatchNorm2d(nn.Module): 7 | """ 8 | BatchNorm2d where the batch statistics and the affine parameters 9 | are fixed 10 | """ 11 | 12 | def __init__(self, n): 13 | super(FrozenBatchNorm2d, self).__init__() 14 | self.register_buffer("weight", torch.ones(n)) 15 | self.register_buffer("bias", torch.zeros(n)) 16 | self.register_buffer("running_mean", torch.zeros(n)) 17 | self.register_buffer("running_var", torch.ones(n)) 18 | 19 | def forward(self, x): 20 | scale = self.weight * self.running_var.rsqrt() 21 | bias = self.bias - self.running_mean * scale 22 | scale = scale.reshape(1, -1, 1, 1) 23 | bias = bias.reshape(1, -1, 1, 1) 24 | return x * scale + bias 25 | -------------------------------------------------------------------------------- /configs/caffe2/e2e_faster_rcnn_X_101_32x8d_FPN_1x_caffe2.yaml: -------------------------------------------------------------------------------- 1 | MODEL: 2 | META_ARCHITECTURE: "GeneralizedRCNN" 3 | WEIGHT: "catalog://Caffe2Detectron/COCO/36761737/e2e_faster_rcnn_X-101-32x8d-FPN_1x" 4 | BACKBONE: 5 | CONV_BODY: "R-101-FPN" 6 | RESNETS: 7 | BACKBONE_OUT_CHANNELS: 256 8 | STRIDE_IN_1X1: False 9 | NUM_GROUPS: 32 10 | WIDTH_PER_GROUP: 8 11 | RPN: 12 | USE_FPN: True 13 | ANCHOR_STRIDE: (4, 8, 16, 32, 64) 14 | PRE_NMS_TOP_N_TRAIN: 2000 15 | PRE_NMS_TOP_N_TEST: 1000 16 | POST_NMS_TOP_N_TEST: 1000 17 | FPN_POST_NMS_TOP_N_TEST: 1000 18 | ROI_HEADS: 19 | USE_FPN: True 20 | ROI_BOX_HEAD: 21 | POOLER_RESOLUTION: 7 22 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 23 | POOLER_SAMPLING_RATIO: 2 24 | FEATURE_EXTRACTOR: "FPN2MLPFeatureExtractor" 25 | PREDICTOR: "FPNPredictor" 26 | DATASETS: 27 | TEST: ("coco_2014_minival",) 28 | DATALOADER: 29 | SIZE_DIVISIBILITY: 32 30 | -------------------------------------------------------------------------------- /matlab/DenseCRF/examples/SalRefineTest.m: -------------------------------------------------------------------------------- 1 | clear all; close all; 2 | addpath([pwd '\..\..\DenseCRF']) 3 | addpath([pwd '\..\..\DenseCRF\include']) 4 | 5 | image = imread('img_1001.jpg'); 6 | SalImage = im2single(imread('img_1001_stage2.png')); 7 | SalImage = repmat(SalImage, [1 1 3]); 8 | SalImage(:,:,1) = 1-SalImage(:,:,1); 9 | 10 | unary = SalImage; 11 | D = Densecrf(); 12 | D.SetImage(image); 13 | D.SetUnary(unary); 14 | 15 | % Some settings. 16 | % D.gaussian_x_stddev = 3; 17 | % D.gaussian_y_stddev = 3; 18 | % D.gaussian_weight = 0.25; 19 | % 20 | % D.bilateral_x_stddev = 64; 21 | % D.bilateral_y_stddev = 64; 22 | % D.bilateral_r_stddev = 4; 23 | % D.bilateral_g_stddev = 4; 24 | % D.bilateral_b_stddev = 4; 25 | % D.bilateral_weight = 0.25; 26 | 27 | D.mean_field; 28 | D.optimization_time 29 | Segmap = D.segmentation-1; 30 | figure;imshow(Segmap) 31 | 32 | D.mean_field_inferprob; 33 | D.optimization_time 34 | Prob = D.prob; 35 | figure;imshow(Prob) -------------------------------------------------------------------------------- /maskrcnn_benchmark/data/transforms/build.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. 2 | from . import transforms as T 3 | 4 | 5 | def build_transforms(cfg, is_train=True): 6 | if is_train: 7 | min_size = cfg.INPUT.MIN_SIZE_TRAIN 8 | max_size = cfg.INPUT.MAX_SIZE_TRAIN 9 | flip_prob = 0.5 # cfg.INPUT.FLIP_PROB_TRAIN 10 | else: 11 | min_size = cfg.INPUT.MIN_SIZE_TEST 12 | max_size = cfg.INPUT.MAX_SIZE_TEST 13 | flip_prob = 0 14 | 15 | to_bgr255 = cfg.INPUT.TO_BGR255 16 | normalize_transform = T.Normalize( 17 | mean=cfg.INPUT.PIXEL_MEAN, std=cfg.INPUT.PIXEL_STD, to_bgr255=to_bgr255 18 | ) 19 | 20 | transform = T.Compose( 21 | [ 22 | T.Resize(min_size, max_size), 23 | T.RandomHorizontalFlip(flip_prob), 24 | T.ToTensor(), 25 | normalize_transform, 26 | ] 27 | ) 28 | return transform 29 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/Cholesky: -------------------------------------------------------------------------------- 1 | #ifndef EIGEN_CHOLESKY_MODULE_H 2 | #define EIGEN_CHOLESKY_MODULE_H 3 | 4 | #include "Core" 5 | 6 | #include "src/Core/util/DisableStupidWarnings.h" 7 | 8 | /** \defgroup Cholesky_Module Cholesky module 9 | * 10 | * 11 | * 12 | * This module provides two variants of the Cholesky decomposition for selfadjoint (hermitian) matrices. 13 | * Those decompositions are accessible via the following MatrixBase methods: 14 | * - MatrixBase::llt(), 15 | * - MatrixBase::ldlt() 16 | * 17 | * \code 18 | * #include 19 | * \endcode 20 | */ 21 | 22 | #include "src/misc/Solve.h" 23 | #include "src/Cholesky/LLT.h" 24 | #include "src/Cholesky/LDLT.h" 25 | #ifdef EIGEN_USE_LAPACKE 26 | #include "src/Cholesky/LLT_MKL.h" 27 | #endif 28 | 29 | #include "src/Core/util/ReenableStupidWarnings.h" 30 | 31 | #endif // EIGEN_CHOLESKY_MODULE_H 32 | /* vim: set filetype=cpp et sw=2 ts=2 ai: */ 33 | -------------------------------------------------------------------------------- /maskrcnn_benchmark/utils/imports.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. 2 | import torch 3 | 4 | if torch._six.PY3: 5 | import importlib 6 | import importlib.util 7 | import sys 8 | 9 | 10 | # from https://stackoverflow.com/questions/67631/how-to-import-a-module-given-the-full-path?utm_medium=organic&utm_source=google_rich_qa&utm_campaign=google_rich_qa 11 | def import_file(module_name, file_path, make_importable=False): 12 | spec = importlib.util.spec_from_file_location(module_name, file_path) 13 | module = importlib.util.module_from_spec(spec) 14 | spec.loader.exec_module(module) 15 | if make_importable: 16 | sys.modules[module_name] = module 17 | return module 18 | else: 19 | import imp 20 | 21 | def import_file(module_name, file_path, make_importable=None): 22 | module = imp.load_source(module_name, file_path) 23 | return module 24 | -------------------------------------------------------------------------------- /matlab/DenseCRF/examples/MyRun.m: -------------------------------------------------------------------------------- 1 | clear all; close all; 2 | addpath('..'); 3 | 4 | image = imread('../include/densecrf/examples/im2.ppm'); 5 | unary = -single(image); 6 | D = Densecrf(); 7 | D.SetImage(image); 8 | D.SetUnary(unary); 9 | D.iterations = 10; 10 | % Some settings. 11 | D.gaussian_x_stddev = 3; 12 | D.gaussian_y_stddev = 3; 13 | D.gaussian_weight = 2^-5; 14 | 15 | D.bilateral_x_stddev = 64; 16 | D.bilateral_y_stddev = 64; 17 | D.bilateral_r_stddev = 4; 18 | D.bilateral_g_stddev = 4; 19 | D.bilateral_b_stddev = 4; 20 | D.bilateral_weight = 2^-5; 21 | 22 | D.mean_field; 23 | D.optimization_time 24 | 25 | 26 | %% Some settings. 27 | D.gaussian_x_stddev = 3; 28 | D.gaussian_y_stddev = 3; 29 | D.gaussian_weight = 2^-5; 30 | 31 | D.bilateral_x_stddev = 60; 32 | D.bilateral_y_stddev = 60; 33 | D.bilateral_r_stddev = 10; 34 | D.bilateral_g_stddev = 10; 35 | D.bilateral_b_stddev = 10; 36 | D.bilateral_weight = 2^-5; 37 | D.mean_field; 38 | D.optimization_time 39 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/decompress_mex.cpp: -------------------------------------------------------------------------------- 1 | // Decompress probabilities 2 | #include "mex.h" 3 | #include "mexutils.h" 4 | #include "cppmatrix.h" 5 | #include "probImage/probimage.h" 6 | 7 | void mexFunction(int nlhs, /* number of expected outputs */ 8 | mxArray *plhs[], /* mxArray output pointer array */ 9 | int nrhs, /* number of inputs */ 10 | const mxArray *prhs[] /* mxArray input pointer array */) 11 | { 12 | char *path = mxArrayToString(prhs[0]); 13 | 14 | ProbImage im; 15 | im.decompress(path); 16 | 17 | const float * data = im.data(); 18 | 19 | matrix unary(im.height(), 20 | im.width(), 21 | im.depth()); 22 | 23 | 24 | for (int x = 0; x < im.width(); x++) { 25 | for (int y = 0; y < im.height(); y++) { 26 | for (int z = 0; z < im.depth(); z++) { 27 | unary(y,x,z) = im(x,y,z); 28 | } 29 | } 30 | } 31 | 32 | plhs[0] = unary; 33 | } -------------------------------------------------------------------------------- /configs/e2e_faster_rcnn_R_50_FPN_1x.yaml: -------------------------------------------------------------------------------- 1 | MODEL: 2 | META_ARCHITECTURE: "GeneralizedRCNN" 3 | WEIGHT: "catalog://ImageNetPretrained/MSRA/R-50" 4 | BACKBONE: 5 | CONV_BODY: "R-50-FPN" 6 | RESNETS: 7 | BACKBONE_OUT_CHANNELS: 256 8 | RPN: 9 | USE_FPN: True 10 | ANCHOR_STRIDE: (4, 8, 16, 32, 64) 11 | PRE_NMS_TOP_N_TRAIN: 2000 12 | PRE_NMS_TOP_N_TEST: 1000 13 | POST_NMS_TOP_N_TEST: 1000 14 | FPN_POST_NMS_TOP_N_TEST: 1000 15 | ROI_HEADS: 16 | USE_FPN: True 17 | ROI_BOX_HEAD: 18 | POOLER_RESOLUTION: 7 19 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 20 | POOLER_SAMPLING_RATIO: 2 21 | FEATURE_EXTRACTOR: "FPN2MLPFeatureExtractor" 22 | PREDICTOR: "FPNPredictor" 23 | DATASETS: 24 | TRAIN: ("coco_2014_train", "coco_2014_valminusminival") 25 | TEST: ("coco_2014_minival",) 26 | DATALOADER: 27 | SIZE_DIVISIBILITY: 32 28 | SOLVER: 29 | BASE_LR: 0.02 30 | WEIGHT_DECAY: 0.0001 31 | STEPS: (60000, 80000) 32 | MAX_ITER: 90000 33 | -------------------------------------------------------------------------------- /configs/e2e_faster_rcnn_R_101_FPN_1x.yaml: -------------------------------------------------------------------------------- 1 | MODEL: 2 | META_ARCHITECTURE: "GeneralizedRCNN" 3 | WEIGHT: "catalog://ImageNetPretrained/MSRA/R-101" 4 | BACKBONE: 5 | CONV_BODY: "R-101-FPN" 6 | RESNETS: 7 | BACKBONE_OUT_CHANNELS: 256 8 | RPN: 9 | USE_FPN: True 10 | ANCHOR_STRIDE: (4, 8, 16, 32, 64) 11 | PRE_NMS_TOP_N_TRAIN: 2000 12 | PRE_NMS_TOP_N_TEST: 1000 13 | POST_NMS_TOP_N_TEST: 1000 14 | FPN_POST_NMS_TOP_N_TEST: 1000 15 | ROI_HEADS: 16 | USE_FPN: True 17 | ROI_BOX_HEAD: 18 | POOLER_RESOLUTION: 7 19 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 20 | POOLER_SAMPLING_RATIO: 2 21 | FEATURE_EXTRACTOR: "FPN2MLPFeatureExtractor" 22 | PREDICTOR: "FPNPredictor" 23 | DATASETS: 24 | TRAIN: ("coco_2014_train", "coco_2014_valminusminival") 25 | TEST: ("coco_2014_minival",) 26 | DATALOADER: 27 | SIZE_DIVISIBILITY: 32 28 | SOLVER: 29 | BASE_LR: 0.02 30 | WEIGHT_DECAY: 0.0001 31 | STEPS: (60000, 80000) 32 | MAX_ITER: 90000 33 | -------------------------------------------------------------------------------- /tests/utils.py: -------------------------------------------------------------------------------- 1 | from __future__ import absolute_import, division, print_function, unicode_literals 2 | 3 | # Set up custom environment before nearly anything else is imported 4 | # NOTE: this should be the first import (no not reorder) 5 | from maskrcnn_benchmark.utils.env import setup_environment # noqa F401 isort:skip 6 | import env_tests.env as env_tests 7 | 8 | import os 9 | import copy 10 | 11 | from maskrcnn_benchmark.config import cfg as g_cfg 12 | 13 | 14 | def get_config_root_path(): 15 | return env_tests.get_config_root_path() 16 | 17 | 18 | def load_config(rel_path): 19 | ''' Load config from file path specified as path relative to config_root ''' 20 | cfg_path = os.path.join(env_tests.get_config_root_path(), rel_path) 21 | return load_config_from_file(cfg_path) 22 | 23 | 24 | def load_config_from_file(file_path): 25 | ''' Load config from file path specified as absolute path ''' 26 | ret = copy.deepcopy(g_cfg) 27 | ret.merge_from_file(file_path) 28 | return ret 29 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/PardisoSupport: -------------------------------------------------------------------------------- 1 | #ifndef EIGEN_PARDISOSUPPORT_MODULE_H 2 | #define EIGEN_PARDISOSUPPORT_MODULE_H 3 | 4 | #include "SparseCore" 5 | 6 | #include "src/Core/util/DisableStupidWarnings.h" 7 | 8 | #include 9 | 10 | #include 11 | 12 | /** \ingroup Support_modules 13 | * \defgroup PardisoSupport_Module PardisoSupport module 14 | * 15 | * This module brings support for the Intel(R) MKL PARDISO direct sparse solvers. 16 | * 17 | * \code 18 | * #include 19 | * \endcode 20 | * 21 | * In order to use this module, the MKL headers must be accessible from the include paths, and your binary must be linked to the MKL library and its dependencies. 22 | * See this \ref TopicUsingIntelMKL "page" for more information on MKL-Eigen integration. 23 | * 24 | */ 25 | 26 | #include "src/PardisoSupport/PardisoSupport.h" 27 | 28 | #include "src/Core/util/ReenableStupidWarnings.h" 29 | 30 | #endif // EIGEN_PARDISOSUPPORT_MODULE_H 31 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/probImage/probimage.h: -------------------------------------------------------------------------------- 1 | #pragma once 2 | 3 | class ProbImage{ 4 | protected: 5 | float * data_; 6 | int width_, height_, depth_; 7 | public: 8 | ProbImage(); 9 | ProbImage( const ProbImage & o ); 10 | ~ProbImage(); 11 | //ProbImage & operator=( const ProbImage & o ); 12 | 13 | // Load and save 14 | void load( const char * file ); 15 | void save( const char * file ); 16 | void compress( const char * file, float eps ); 17 | void decompress( const char * file ); 18 | 19 | // Properties 20 | int width() const { return width_; } 21 | int height() const { return height_; } 22 | int depth() const { return depth_; } 23 | 24 | // Conversion operations 25 | void boostToProb(); 26 | void probToBoost(); 27 | 28 | // Data access 29 | const float & operator()( int i, int j, int k ) const { return data_[(j*width_+i)*depth_+k]; } 30 | float & operator()( int i, int j, int k ) { return data_[(j*width_+i)*depth_+k]; } 31 | const float * data() const { return data_; } 32 | float * data() { return data_; } 33 | }; 34 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/SparseCholesky: -------------------------------------------------------------------------------- 1 | #ifndef EIGEN_SPARSECHOLESKY_MODULE_H 2 | #define EIGEN_SPARSECHOLESKY_MODULE_H 3 | 4 | #include "SparseCore" 5 | 6 | #include "src/Core/util/DisableStupidWarnings.h" 7 | 8 | /** \ingroup Sparse_modules 9 | * \defgroup SparseCholesky_Module SparseCholesky module 10 | * 11 | * This module currently provides two variants of the direct sparse Cholesky decomposition for selfadjoint (hermitian) matrices. 12 | * Those decompositions are accessible via the following classes: 13 | * - SimplicialLLt, 14 | * - SimplicialLDLt 15 | * 16 | * Such problems can also be solved using the ConjugateGradient solver from the IterativeLinearSolvers module. 17 | * 18 | * \code 19 | * #include 20 | * \endcode 21 | */ 22 | 23 | #include "src/misc/Solve.h" 24 | #include "src/misc/SparseSolve.h" 25 | 26 | #include "src/SparseCholesky/SimplicialCholesky.h" 27 | 28 | #include "src/Core/util/ReenableStupidWarnings.h" 29 | 30 | #endif // EIGEN_SPARSECHOLESKY_MODULE_H 31 | -------------------------------------------------------------------------------- /configs/cityscapes/e2e_faster_rcnn_R_50_FPN_1x_cocostyle.yaml: -------------------------------------------------------------------------------- 1 | MODEL: 2 | META_ARCHITECTURE: "GeneralizedRCNN" 3 | WEIGHT: "catalog://ImageNetPretrained/MSRA/R-50" 4 | BACKBONE: 5 | CONV_BODY: "R-50-FPN" 6 | RESNETS: 7 | BACKBONE_OUT_CHANNELS: 256 8 | RPN: 9 | USE_FPN: True 10 | ANCHOR_STRIDE: (4, 8, 16, 32, 64) 11 | PRE_NMS_TOP_N_TRAIN: 2000 12 | PRE_NMS_TOP_N_TEST: 1000 13 | POST_NMS_TOP_N_TEST: 1000 14 | FPN_POST_NMS_TOP_N_TEST: 1000 15 | ROI_HEADS: 16 | USE_FPN: True 17 | ROI_BOX_HEAD: 18 | POOLER_RESOLUTION: 7 19 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 20 | POOLER_SAMPLING_RATIO: 2 21 | FEATURE_EXTRACTOR: "FPN2MLPFeatureExtractor" 22 | PREDICTOR: "FPNPredictor" 23 | NUM_CLASSES: 9 24 | DATASETS: 25 | TRAIN: ("cityscapes_fine_instanceonly_seg_train_cocostyle",) 26 | TEST: ("cityscapes_fine_instanceonly_seg_val_cocostyle",) 27 | DATALOADER: 28 | SIZE_DIVISIBILITY: 32 29 | SOLVER: 30 | BASE_LR: 0.01 31 | WEIGHT_DECAY: 0.0001 32 | STEPS: (18000,) 33 | MAX_ITER: 24000 34 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/SVD: -------------------------------------------------------------------------------- 1 | #ifndef EIGEN_SVD_MODULE_H 2 | #define EIGEN_SVD_MODULE_H 3 | 4 | #include "QR" 5 | #include "Householder" 6 | #include "Jacobi" 7 | 8 | #include "src/Core/util/DisableStupidWarnings.h" 9 | 10 | /** \defgroup SVD_Module SVD module 11 | * 12 | * 13 | * 14 | * This module provides SVD decomposition for matrices (both real and complex). 15 | * This decomposition is accessible via the following MatrixBase method: 16 | * - MatrixBase::jacobiSvd() 17 | * 18 | * \code 19 | * #include 20 | * \endcode 21 | */ 22 | 23 | #include "src/misc/Solve.h" 24 | #include "src/SVD/JacobiSVD.h" 25 | #if defined(EIGEN_USE_LAPACKE) && !defined(EIGEN_USE_LAPACKE_STRICT) 26 | #include "src/SVD/JacobiSVD_MKL.h" 27 | #endif 28 | #include "src/SVD/UpperBidiagonalization.h" 29 | 30 | #ifdef EIGEN2_SUPPORT 31 | #include "src/Eigen2Support/SVD.h" 32 | #endif 33 | 34 | #include "src/Core/util/ReenableStupidWarnings.h" 35 | 36 | #endif // EIGEN_SVD_MODULE_H 37 | /* vim: set filetype=cpp et sw=2 ts=2 ai: */ 38 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/extendedDenseCRF2D.h: -------------------------------------------------------------------------------- 1 | #pragma once 2 | 3 | #include 4 | #include 5 | #include "densecrf/include/densecrf.h" 6 | 7 | // Missing in densecrf.h 8 | void expAndNormalize ( Eigen::MatrixXf & out, const Eigen::MatrixXf & in ); 9 | 10 | class extendedDenseCRF2D : public DenseCRF2D 11 | { 12 | public: 13 | extendedDenseCRF2D( int W, int H, int M) : DenseCRF2D(W,H,M) 14 | {} 15 | 16 | double energy(VectorXs map) 17 | { 18 | Eigen::VectorXf unary = unaryEnergy( map ); 19 | Eigen::VectorXf pairwise = pairwiseEnergy( map ); 20 | 21 | double energy = 0; 22 | 23 | for (int i = 0; i < W_*H_; i++) 24 | energy += unary(i); 25 | 26 | for (int i = 0; i < W_*H_; i++) 27 | energy -= pairwise(i); 28 | 29 | // Correct for negative weighted potts. 30 | // the pairwise cost is -w[a==b], but the paper states w[a!=b]. 31 | map *= 0; 32 | pairwise = pairwiseEnergy( map ); 33 | 34 | for (int i = 0; i < W_*H_; i++) 35 | energy += pairwise(i); 36 | 37 | return energy; 38 | } 39 | }; -------------------------------------------------------------------------------- /matlab/DenseCRF/include/TRW_S-v1.3/README.TXT: -------------------------------------------------------------------------------- 1 | This software implements two algorithms for minimizing energy functions of the form 2 | 3 | E(x) = \sum_i D_i(x_i) + \sum_ij V_ij(x_i,x_j) 4 | where x_i are discrete variables. 5 | 6 | The two algorithms are max-product belief propagation (BP, Pearl'88) and 7 | sequential tree-reweighted max-product message passing (TRW-S, Kolmogorov'05). 8 | 9 | For example usage look at one of the type*.h files 10 | (typeBinary.h, typeBinaryFast.h, typePotts.h, typeGeneral.h, 11 | typeTruncatedLinear.h, typeTruncatedQuadratic.h, typeTruncatedLinear2D.h, typeTruncatedQuadratic2D.h). 12 | If your energy function does not belong to the classes defined 13 | in these files but terms V_ij allow fast distance transforms, then 14 | it should be possible to extend the algorithms to your functions 15 | using files type*.h as examples. 16 | 17 | Written by Vladimir Kolmogorov (vnk@microsoft.com), 2005. 18 | Tested under Microsoft Visual Studio .NET (Windows) 19 | and GNU c++ compiler version 2.96 (Red Hat Linux 7.1). 20 | 21 | (c) Microsoft Corporation. All rights reserved. 22 | -------------------------------------------------------------------------------- /configs/e2e_faster_rcnn_X_101_32x8d_FPN_1x.yaml: -------------------------------------------------------------------------------- 1 | MODEL: 2 | META_ARCHITECTURE: "GeneralizedRCNN" 3 | WEIGHT: "catalog://ImageNetPretrained/FAIR/20171220/X-101-32x8d" 4 | BACKBONE: 5 | CONV_BODY: "R-101-FPN" 6 | RPN: 7 | USE_FPN: True 8 | ANCHOR_STRIDE: (4, 8, 16, 32, 64) 9 | PRE_NMS_TOP_N_TRAIN: 2000 10 | PRE_NMS_TOP_N_TEST: 1000 11 | POST_NMS_TOP_N_TEST: 1000 12 | FPN_POST_NMS_TOP_N_TEST: 1000 13 | ROI_HEADS: 14 | USE_FPN: True 15 | ROI_BOX_HEAD: 16 | POOLER_RESOLUTION: 7 17 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 18 | POOLER_SAMPLING_RATIO: 2 19 | FEATURE_EXTRACTOR: "FPN2MLPFeatureExtractor" 20 | PREDICTOR: "FPNPredictor" 21 | RESNETS: 22 | BACKBONE_OUT_CHANNELS: 256 23 | STRIDE_IN_1X1: False 24 | NUM_GROUPS: 32 25 | WIDTH_PER_GROUP: 8 26 | DATASETS: 27 | TRAIN: ("coco_2014_train", "coco_2014_valminusminival") 28 | TEST: ("coco_2014_minival",) 29 | DATALOADER: 30 | SIZE_DIVISIBILITY: 32 31 | SOLVER: 32 | BASE_LR: 0.01 33 | WEIGHT_DECAY: 0.0001 34 | STEPS: (120000, 160000) 35 | MAX_ITER: 180000 36 | IMS_PER_BATCH: 8 37 | -------------------------------------------------------------------------------- /maskrcnn_benchmark/solver/build.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. 2 | import torch 3 | 4 | from .lr_scheduler import WarmupMultiStepLR 5 | 6 | 7 | def make_optimizer(cfg, model): 8 | params = [] 9 | for key, value in model.named_parameters(): 10 | if not value.requires_grad: 11 | continue 12 | lr = cfg.SOLVER.BASE_LR 13 | weight_decay = cfg.SOLVER.WEIGHT_DECAY 14 | if "bias" in key: 15 | lr = cfg.SOLVER.BASE_LR * cfg.SOLVER.BIAS_LR_FACTOR 16 | weight_decay = cfg.SOLVER.WEIGHT_DECAY_BIAS 17 | params += [{"params": [value], "lr": lr, "weight_decay": weight_decay}] 18 | 19 | optimizer = torch.optim.SGD(params, lr, momentum=cfg.SOLVER.MOMENTUM) 20 | return optimizer 21 | 22 | 23 | def make_lr_scheduler(cfg, optimizer): 24 | return WarmupMultiStepLR( 25 | optimizer, 26 | cfg.SOLVER.STEPS, 27 | cfg.SOLVER.GAMMA, 28 | warmup_factor=cfg.SOLVER.WARMUP_FACTOR, 29 | warmup_iters=cfg.SOLVER.WARMUP_ITERS, 30 | warmup_method=cfg.SOLVER.WARMUP_METHOD, 31 | ) 32 | -------------------------------------------------------------------------------- /maskrcnn_benchmark/data/datasets/list_dataset.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. 2 | """ 3 | Simple dataset class that wraps a list of path names 4 | """ 5 | 6 | from PIL import Image 7 | 8 | from maskrcnn_benchmark.structures.bounding_box import BoxList 9 | 10 | 11 | class ListDataset(object): 12 | def __init__(self, image_lists, transforms=None): 13 | self.image_lists = image_lists 14 | self.transforms = transforms 15 | 16 | def __getitem__(self, item): 17 | img = Image.open(self.image_lists[item]).convert("RGB") 18 | 19 | # dummy target 20 | w, h = img.size 21 | target = BoxList([[0, 0, w, h]], img.size, mode="xyxy") 22 | 23 | if self.transforms is not None: 24 | img, target = self.transforms(img, target) 25 | 26 | return img, target 27 | 28 | def __len__(self): 29 | return len(self.image_lists) 30 | 31 | def get_img_info(self, item): 32 | """ 33 | Return the image dimensions for the image, without 34 | loading and pre-processing it 35 | """ 36 | pass 37 | -------------------------------------------------------------------------------- /LICENSE: -------------------------------------------------------------------------------- 1 | MIT License 2 | 3 | Copyright (c) 2019 chengchun.hsu 4 | 5 | Permission is hereby granted, free of charge, to any person obtaining a copy 6 | of this software and associated documentation files (the "Software"), to deal 7 | in the Software without restriction, including without limitation the rights 8 | to use, copy, modify, merge, publish, distribute, sublicense, and/or sell 9 | copies of the Software, and to permit persons to whom the Software is 10 | furnished to do so, subject to the following conditions: 11 | 12 | The above copyright notice and this permission notice shall be included in all 13 | copies or substantial portions of the Software. 14 | 15 | THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR 16 | IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, 17 | FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE 18 | AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER 19 | LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, 20 | OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE 21 | SOFTWARE. 22 | -------------------------------------------------------------------------------- /maskrcnn_benchmark/data/datasets/evaluation/__init__.py: -------------------------------------------------------------------------------- 1 | from maskrcnn_benchmark.data import datasets 2 | 3 | from .coco import coco_evaluation 4 | from .voc import voc_evaluation 5 | 6 | 7 | def evaluate(dataset, predictions, output_folder, **kwargs): 8 | """evaluate dataset using different methods based on dataset type. 9 | Args: 10 | dataset: Dataset object 11 | predictions(list[BoxList]): each item in the list represents the 12 | prediction results for one image. 13 | output_folder: output folder, to save evaluation files or results. 14 | **kwargs: other args. 15 | Returns: 16 | evaluation result 17 | """ 18 | args = dict( 19 | dataset=dataset, predictions=predictions, output_folder=output_folder, **kwargs 20 | ) 21 | if isinstance(dataset, datasets.COCODataset): 22 | return coco_evaluation(**args) 23 | elif isinstance(dataset, datasets.PascalVOCDataset): 24 | return voc_evaluation(**args) 25 | else: 26 | dataset_name = dataset.__class__.__name__ 27 | raise NotImplementedError("Unsupported dataset type {}.".format(dataset_name)) 28 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/src/SparseCore/SparseFuzzy.h: -------------------------------------------------------------------------------- 1 | // This file is part of Eigen, a lightweight C++ template library 2 | // for linear algebra. 3 | // 4 | // Copyright (C) 2008 Gael Guennebaud 5 | // 6 | // This Source Code Form is subject to the terms of the Mozilla 7 | // Public License v. 2.0. If a copy of the MPL was not distributed 8 | // with this file, You can obtain one at http://mozilla.org/MPL/2.0/. 9 | 10 | #ifndef EIGEN_SPARSE_FUZZY_H 11 | #define EIGEN_SPARSE_FUZZY_H 12 | 13 | // template 14 | // template 15 | // bool SparseMatrixBase::isApprox( 16 | // const OtherDerived& other, 17 | // typename NumTraits::Real prec 18 | // ) const 19 | // { 20 | // const typename internal::nested::type nested(derived()); 21 | // const typename internal::nested::type otherNested(other.derived()); 22 | // return (nested - otherNested).cwise().abs2().sum() 23 | // <= prec * prec * (std::min)(nested.cwise().abs2().sum(), otherNested.cwise().abs2().sum()); 24 | // } 25 | 26 | #endif // EIGEN_SPARSE_FUZZY_H 27 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/QR: -------------------------------------------------------------------------------- 1 | #ifndef EIGEN_QR_MODULE_H 2 | #define EIGEN_QR_MODULE_H 3 | 4 | #include "Core" 5 | 6 | #include "src/Core/util/DisableStupidWarnings.h" 7 | 8 | #include "Cholesky" 9 | #include "Jacobi" 10 | #include "Householder" 11 | 12 | /** \defgroup QR_Module QR module 13 | * 14 | * 15 | * 16 | * This module provides various QR decompositions 17 | * This module also provides some MatrixBase methods, including: 18 | * - MatrixBase::qr(), 19 | * 20 | * \code 21 | * #include 22 | * \endcode 23 | */ 24 | 25 | #include "src/misc/Solve.h" 26 | #include "src/QR/HouseholderQR.h" 27 | #include "src/QR/FullPivHouseholderQR.h" 28 | #include "src/QR/ColPivHouseholderQR.h" 29 | #ifdef EIGEN_USE_LAPACKE 30 | #include "src/QR/HouseholderQR_MKL.h" 31 | #include "src/QR/ColPivHouseholderQR_MKL.h" 32 | #endif 33 | 34 | #ifdef EIGEN2_SUPPORT 35 | #include "src/Eigen2Support/QR.h" 36 | #endif 37 | 38 | #include "src/Core/util/ReenableStupidWarnings.h" 39 | 40 | #ifdef EIGEN2_SUPPORT 41 | #include "Eigenvalues" 42 | #endif 43 | 44 | #endif // EIGEN_QR_MODULE_H 45 | /* vim: set filetype=cpp et sw=2 ts=2 ai: */ 46 | -------------------------------------------------------------------------------- /configs/caffe2/e2e_mask_rcnn_R_50_FPN_1x_caffe2.yaml: -------------------------------------------------------------------------------- 1 | MODEL: 2 | META_ARCHITECTURE: "GeneralizedRCNN" 3 | WEIGHT: "catalog://Caffe2Detectron/COCO/35858933/e2e_mask_rcnn_R-50-FPN_1x" 4 | BACKBONE: 5 | CONV_BODY: "R-50-FPN" 6 | RESNETS: 7 | BACKBONE_OUT_CHANNELS: 256 8 | RPN: 9 | USE_FPN: True 10 | ANCHOR_STRIDE: (4, 8, 16, 32, 64) 11 | PRE_NMS_TOP_N_TRAIN: 2000 12 | PRE_NMS_TOP_N_TEST: 1000 13 | POST_NMS_TOP_N_TEST: 1000 14 | FPN_POST_NMS_TOP_N_TEST: 1000 15 | ROI_HEADS: 16 | USE_FPN: True 17 | ROI_BOX_HEAD: 18 | POOLER_RESOLUTION: 7 19 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 20 | POOLER_SAMPLING_RATIO: 2 21 | FEATURE_EXTRACTOR: "FPN2MLPFeatureExtractor" 22 | PREDICTOR: "FPNPredictor" 23 | ROI_MASK_HEAD: 24 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 25 | FEATURE_EXTRACTOR: "MaskRCNNFPNFeatureExtractor" 26 | PREDICTOR: "MaskRCNNC4Predictor" 27 | POOLER_RESOLUTION: 14 28 | POOLER_SAMPLING_RATIO: 2 29 | RESOLUTION: 28 30 | SHARE_BOX_FEATURE_EXTRACTOR: False 31 | MASK_ON: True 32 | DATASETS: 33 | TEST: ("coco_2014_minival",) 34 | DATALOADER: 35 | SIZE_DIVISIBILITY: 32 36 | -------------------------------------------------------------------------------- /matlab/DenseCRF/LICENSE.md: -------------------------------------------------------------------------------- 1 | The MIT License (MIT) 2 | 3 | Copyright (c) 2015 Johannes Ulén 4 | 5 | Permission is hereby granted, free of charge, to any person obtaining a copy 6 | of this software and associated documentation files (the "Software"), to deal 7 | in the Software without restriction, including without limitation the rights 8 | to use, copy, modify, merge, publish, distribute, sublicense, and/or sell 9 | copies of the Software, and to permit persons to whom the Software is 10 | furnished to do so, subject to the following conditions: 11 | 12 | The above copyright notice and this permission notice shall be included in all 13 | copies or substantial portions of the Software. 14 | 15 | THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR 16 | IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, 17 | FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE 18 | AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER 19 | LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, 20 | OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE 21 | SOFTWARE. 22 | -------------------------------------------------------------------------------- /configs/caffe2/e2e_mask_rcnn_R_101_FPN_1x_caffe2.yaml: -------------------------------------------------------------------------------- 1 | MODEL: 2 | META_ARCHITECTURE: "GeneralizedRCNN" 3 | WEIGHT: "catalog://Caffe2Detectron/COCO/35861795/e2e_mask_rcnn_R-101-FPN_1x" 4 | BACKBONE: 5 | CONV_BODY: "R-101-FPN" 6 | RESNETS: 7 | BACKBONE_OUT_CHANNELS: 256 8 | RPN: 9 | USE_FPN: True 10 | ANCHOR_STRIDE: (4, 8, 16, 32, 64) 11 | PRE_NMS_TOP_N_TRAIN: 2000 12 | PRE_NMS_TOP_N_TEST: 1000 13 | POST_NMS_TOP_N_TEST: 1000 14 | FPN_POST_NMS_TOP_N_TEST: 1000 15 | ROI_HEADS: 16 | USE_FPN: True 17 | ROI_BOX_HEAD: 18 | POOLER_RESOLUTION: 7 19 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 20 | POOLER_SAMPLING_RATIO: 2 21 | FEATURE_EXTRACTOR: "FPN2MLPFeatureExtractor" 22 | PREDICTOR: "FPNPredictor" 23 | ROI_MASK_HEAD: 24 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 25 | FEATURE_EXTRACTOR: "MaskRCNNFPNFeatureExtractor" 26 | PREDICTOR: "MaskRCNNC4Predictor" 27 | POOLER_RESOLUTION: 14 28 | POOLER_SAMPLING_RATIO: 2 29 | RESOLUTION: 28 30 | SHARE_BOX_FEATURE_EXTRACTOR: False 31 | MASK_ON: True 32 | DATASETS: 33 | TEST: ("coco_2014_minival",) 34 | DATALOADER: 35 | SIZE_DIVISIBILITY: 32 36 | -------------------------------------------------------------------------------- /configs/quick_schedules/e2e_faster_rcnn_R_50_FPN_quick.yaml: -------------------------------------------------------------------------------- 1 | MODEL: 2 | META_ARCHITECTURE: "GeneralizedRCNN" 3 | WEIGHT: "catalog://ImageNetPretrained/MSRA/R-50" 4 | BACKBONE: 5 | CONV_BODY: "R-50-FPN" 6 | RESNETS: 7 | BACKBONE_OUT_CHANNELS: 256 8 | RPN: 9 | USE_FPN: True 10 | ANCHOR_STRIDE: (4, 8, 16, 32, 64) 11 | PRE_NMS_TOP_N_TRAIN: 2000 12 | PRE_NMS_TOP_N_TEST: 1000 13 | POST_NMS_TOP_N_TEST: 1000 14 | FPN_POST_NMS_TOP_N_TEST: 1000 15 | ROI_HEADS: 16 | USE_FPN: True 17 | BATCH_SIZE_PER_IMAGE: 256 18 | ROI_BOX_HEAD: 19 | POOLER_RESOLUTION: 7 20 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 21 | POOLER_SAMPLING_RATIO: 2 22 | FEATURE_EXTRACTOR: "FPN2MLPFeatureExtractor" 23 | PREDICTOR: "FPNPredictor" 24 | DATASETS: 25 | TRAIN: ("coco_2014_minival",) 26 | TEST: ("coco_2014_minival",) 27 | INPUT: 28 | MIN_SIZE_TRAIN: (600,) 29 | MAX_SIZE_TRAIN: 1000 30 | MIN_SIZE_TEST: 800 31 | MAX_SIZE_TEST: 1000 32 | DATALOADER: 33 | SIZE_DIVISIBILITY: 32 34 | SOLVER: 35 | BASE_LR: 0.005 36 | WEIGHT_DECAY: 0.0001 37 | STEPS: (1500,) 38 | MAX_ITER: 2000 39 | IMS_PER_BATCH: 4 40 | TEST: 41 | IMS_PER_BATCH: 2 42 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/LU: -------------------------------------------------------------------------------- 1 | #ifndef EIGEN_LU_MODULE_H 2 | #define EIGEN_LU_MODULE_H 3 | 4 | #include "Core" 5 | 6 | #include "src/Core/util/DisableStupidWarnings.h" 7 | 8 | /** \defgroup LU_Module LU module 9 | * This module includes %LU decomposition and related notions such as matrix inversion and determinant. 10 | * This module defines the following MatrixBase methods: 11 | * - MatrixBase::inverse() 12 | * - MatrixBase::determinant() 13 | * 14 | * \code 15 | * #include 16 | * \endcode 17 | */ 18 | 19 | #include "src/misc/Solve.h" 20 | #include "src/misc/Kernel.h" 21 | #include "src/misc/Image.h" 22 | #include "src/LU/FullPivLU.h" 23 | #include "src/LU/PartialPivLU.h" 24 | #ifdef EIGEN_USE_LAPACKE 25 | #include "src/LU/PartialPivLU_MKL.h" 26 | #endif 27 | #include "src/LU/Determinant.h" 28 | #include "src/LU/Inverse.h" 29 | 30 | #if defined EIGEN_VECTORIZE_SSE 31 | #include "src/LU/arch/Inverse_SSE.h" 32 | #endif 33 | 34 | #ifdef EIGEN2_SUPPORT 35 | #include "src/Eigen2Support/LU.h" 36 | #endif 37 | 38 | #include "src/Core/util/ReenableStupidWarnings.h" 39 | 40 | #endif // EIGEN_LU_MODULE_H 41 | /* vim: set filetype=cpp et sw=2 ts=2 ai: */ 42 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/external/liblbfgs/COPYING: -------------------------------------------------------------------------------- 1 | The MIT License 2 | 3 | Copyright (c) 1990 Jorge Nocedal 4 | Copyright (c) 2007-2010 Naoaki Okazaki 5 | 6 | Permission is hereby granted, free of charge, to any person obtaining a 7 | copy of this software and associated documentation files (the "Software"), 8 | to deal in the Software without restriction, including without limitation 9 | the rights to use, copy, modify, merge, publish, distribute, sublicense, 10 | and/or sell copies of the Software, and to permit persons to whom the 11 | Software is furnished to do so, subject to the following conditions: 12 | 13 | The above copyright notice and this permission notice shall be included in 14 | all copies or substantial portions of the Software. 15 | 16 | THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR 17 | IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, 18 | FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE 19 | AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER 20 | LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, 21 | OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN 22 | THE SOFTWARE. 23 | -------------------------------------------------------------------------------- /maskrcnn_benchmark/csrc/SigmoidFocalLoss.h: -------------------------------------------------------------------------------- 1 | #pragma once 2 | 3 | #include "cpu/vision.h" 4 | 5 | #ifdef WITH_CUDA 6 | #include "cuda/vision.h" 7 | #endif 8 | 9 | // Interface for Python 10 | at::Tensor SigmoidFocalLoss_forward( 11 | const at::Tensor& logits, 12 | const at::Tensor& targets, 13 | const int num_classes, 14 | const float gamma, 15 | const float alpha) { 16 | if (logits.type().is_cuda()) { 17 | #ifdef WITH_CUDA 18 | return SigmoidFocalLoss_forward_cuda(logits, targets, num_classes, gamma, alpha); 19 | #else 20 | AT_ERROR("Not compiled with GPU support"); 21 | #endif 22 | } 23 | AT_ERROR("Not implemented on the CPU"); 24 | } 25 | 26 | at::Tensor SigmoidFocalLoss_backward( 27 | const at::Tensor& logits, 28 | const at::Tensor& targets, 29 | const at::Tensor& d_losses, 30 | const int num_classes, 31 | const float gamma, 32 | const float alpha) { 33 | if (logits.type().is_cuda()) { 34 | #ifdef WITH_CUDA 35 | return SigmoidFocalLoss_backward_cuda(logits, targets, d_losses, num_classes, gamma, alpha); 36 | #else 37 | AT_ERROR("Not compiled with GPU support"); 38 | #endif 39 | } 40 | AT_ERROR("Not implemented on the CPU"); 41 | } 42 | -------------------------------------------------------------------------------- /configs/e2e_faster_rcnn_fbnet.yaml: -------------------------------------------------------------------------------- 1 | MODEL: 2 | META_ARCHITECTURE: "GeneralizedRCNN" 3 | BACKBONE: 4 | CONV_BODY: FBNet 5 | FBNET: 6 | ARCH: "default" 7 | BN_TYPE: "bn" 8 | WIDTH_DIVISOR: 8 9 | DW_CONV_SKIP_BN: True 10 | DW_CONV_SKIP_RELU: True 11 | RPN: 12 | ANCHOR_SIZES: (16, 32, 64, 128, 256) 13 | ANCHOR_STRIDE: (16, ) 14 | BATCH_SIZE_PER_IMAGE: 256 15 | PRE_NMS_TOP_N_TRAIN: 6000 16 | PRE_NMS_TOP_N_TEST: 6000 17 | POST_NMS_TOP_N_TRAIN: 2000 18 | POST_NMS_TOP_N_TEST: 100 19 | RPN_HEAD: FBNet.rpn_head 20 | ROI_HEADS: 21 | BATCH_SIZE_PER_IMAGE: 512 22 | ROI_BOX_HEAD: 23 | POOLER_RESOLUTION: 6 24 | FEATURE_EXTRACTOR: FBNet.roi_head 25 | NUM_CLASSES: 81 26 | DATASETS: 27 | TRAIN: ("coco_2014_train", "coco_2014_valminusminival") 28 | TEST: ("coco_2014_minival",) 29 | SOLVER: 30 | BASE_LR: 0.06 31 | WARMUP_FACTOR: 0.1 32 | WEIGHT_DECAY: 0.0001 33 | STEPS: (60000, 80000) 34 | MAX_ITER: 90000 35 | IMS_PER_BATCH: 128 # for 8GPUs 36 | # TEST: 37 | # IMS_PER_BATCH: 8 38 | INPUT: 39 | MIN_SIZE_TRAIN: (320, ) 40 | MAX_SIZE_TRAIN: 640 41 | MIN_SIZE_TEST: 320 42 | MAX_SIZE_TEST: 640 43 | PIXEL_MEAN: [103.53, 116.28, 123.675] 44 | PIXEL_STD: [57.375, 57.12, 58.395] 45 | -------------------------------------------------------------------------------- /configs/e2e_faster_rcnn_fbnet_600.yaml: -------------------------------------------------------------------------------- 1 | MODEL: 2 | META_ARCHITECTURE: "GeneralizedRCNN" 3 | BACKBONE: 4 | CONV_BODY: FBNet 5 | FBNET: 6 | ARCH: "default" 7 | BN_TYPE: "bn" 8 | WIDTH_DIVISOR: 8 9 | DW_CONV_SKIP_BN: True 10 | DW_CONV_SKIP_RELU: True 11 | RPN: 12 | ANCHOR_SIZES: (32, 64, 128, 256, 512) 13 | ANCHOR_STRIDE: (16, ) 14 | BATCH_SIZE_PER_IMAGE: 256 15 | PRE_NMS_TOP_N_TRAIN: 6000 16 | PRE_NMS_TOP_N_TEST: 6000 17 | POST_NMS_TOP_N_TRAIN: 2000 18 | POST_NMS_TOP_N_TEST: 200 19 | RPN_HEAD: FBNet.rpn_head 20 | ROI_HEADS: 21 | BATCH_SIZE_PER_IMAGE: 256 22 | ROI_BOX_HEAD: 23 | POOLER_RESOLUTION: 6 24 | FEATURE_EXTRACTOR: FBNet.roi_head 25 | NUM_CLASSES: 81 26 | DATASETS: 27 | TRAIN: ("coco_2014_train", "coco_2014_valminusminival") 28 | TEST: ("coco_2014_minival",) 29 | SOLVER: 30 | BASE_LR: 0.06 31 | WARMUP_FACTOR: 0.1 32 | WEIGHT_DECAY: 0.0001 33 | STEPS: (60000, 80000) 34 | MAX_ITER: 90000 35 | IMS_PER_BATCH: 128 # for 8GPUs 36 | # TEST: 37 | # IMS_PER_BATCH: 8 38 | INPUT: 39 | MIN_SIZE_TRAIN: (600, ) 40 | MAX_SIZE_TRAIN: 1000 41 | MIN_SIZE_TEST: 600 42 | MAX_SIZE_TEST: 1000 43 | PIXEL_MEAN: [103.53, 116.28, 123.675] 44 | PIXEL_STD: [57.375, 57.12, 58.395] 45 | -------------------------------------------------------------------------------- /configs/caffe2/e2e_mask_rcnn_X_101_32x8d_FPN_1x_caffe2.yaml: -------------------------------------------------------------------------------- 1 | MODEL: 2 | META_ARCHITECTURE: "GeneralizedRCNN" 3 | WEIGHT: "catalog://Caffe2Detectron/COCO/36761843/e2e_mask_rcnn_X-101-32x8d-FPN_1x" 4 | BACKBONE: 5 | CONV_BODY: "R-101-FPN" 6 | RESNETS: 7 | BACKBONE_OUT_CHANNELS: 256 8 | STRIDE_IN_1X1: False 9 | NUM_GROUPS: 32 10 | WIDTH_PER_GROUP: 8 11 | RPN: 12 | USE_FPN: True 13 | ANCHOR_STRIDE: (4, 8, 16, 32, 64) 14 | PRE_NMS_TOP_N_TRAIN: 2000 15 | PRE_NMS_TOP_N_TEST: 1000 16 | POST_NMS_TOP_N_TEST: 1000 17 | FPN_POST_NMS_TOP_N_TEST: 1000 18 | ROI_HEADS: 19 | USE_FPN: True 20 | ROI_BOX_HEAD: 21 | POOLER_RESOLUTION: 7 22 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 23 | POOLER_SAMPLING_RATIO: 2 24 | FEATURE_EXTRACTOR: "FPN2MLPFeatureExtractor" 25 | PREDICTOR: "FPNPredictor" 26 | ROI_MASK_HEAD: 27 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 28 | FEATURE_EXTRACTOR: "MaskRCNNFPNFeatureExtractor" 29 | PREDICTOR: "MaskRCNNC4Predictor" 30 | POOLER_RESOLUTION: 14 31 | POOLER_SAMPLING_RATIO: 2 32 | RESOLUTION: 28 33 | SHARE_BOX_FEATURE_EXTRACTOR: False 34 | MASK_ON: True 35 | DATASETS: 36 | TEST: ("coco_2014_minival",) 37 | DATALOADER: 38 | SIZE_DIVISIBILITY: 32 39 | -------------------------------------------------------------------------------- /configs/e2e_faster_rcnn_fbnet_chamv1a_600.yaml: -------------------------------------------------------------------------------- 1 | MODEL: 2 | META_ARCHITECTURE: "GeneralizedRCNN" 3 | BACKBONE: 4 | CONV_BODY: FBNet 5 | FBNET: 6 | ARCH: "cham_v1a" 7 | BN_TYPE: "bn" 8 | WIDTH_DIVISOR: 8 9 | DW_CONV_SKIP_BN: True 10 | DW_CONV_SKIP_RELU: True 11 | RPN: 12 | ANCHOR_SIZES: (32, 64, 128, 256, 512) 13 | ANCHOR_STRIDE: (16, ) 14 | BATCH_SIZE_PER_IMAGE: 256 15 | PRE_NMS_TOP_N_TRAIN: 6000 16 | PRE_NMS_TOP_N_TEST: 6000 17 | POST_NMS_TOP_N_TRAIN: 2000 18 | POST_NMS_TOP_N_TEST: 200 19 | RPN_HEAD: FBNet.rpn_head 20 | ROI_HEADS: 21 | BATCH_SIZE_PER_IMAGE: 128 22 | ROI_BOX_HEAD: 23 | POOLER_RESOLUTION: 6 24 | FEATURE_EXTRACTOR: FBNet.roi_head 25 | NUM_CLASSES: 81 26 | DATASETS: 27 | TRAIN: ("coco_2014_train", "coco_2014_valminusminival") 28 | TEST: ("coco_2014_minival",) 29 | SOLVER: 30 | BASE_LR: 0.045 31 | WARMUP_FACTOR: 0.1 32 | WEIGHT_DECAY: 0.0001 33 | STEPS: (90000, 120000) 34 | MAX_ITER: 135000 35 | IMS_PER_BATCH: 96 # for 8GPUs 36 | # TEST: 37 | # IMS_PER_BATCH: 8 38 | INPUT: 39 | MIN_SIZE_TRAIN: (600, ) 40 | MAX_SIZE_TRAIN: 1000 41 | MIN_SIZE_TEST: 600 42 | MAX_SIZE_TEST: 1000 43 | PIXEL_MEAN: [103.53, 116.28, 123.675] 44 | PIXEL_STD: [57.375, 57.12, 58.395] 45 | -------------------------------------------------------------------------------- /configs/caffe2/e2e_mask_rcnn_X-152-32x8d-FPN-IN5k_1.44x_caffe2.yaml: -------------------------------------------------------------------------------- 1 | MODEL: 2 | META_ARCHITECTURE: "GeneralizedRCNN" 3 | WEIGHT: "catalog://Caffe2Detectron/COCO/37129812/e2e_mask_rcnn_X-152-32x8d-FPN-IN5k_1.44x" 4 | BACKBONE: 5 | CONV_BODY: "R-152-FPN" 6 | RESNETS: 7 | BACKBONE_OUT_CHANNELS: 256 8 | STRIDE_IN_1X1: False 9 | NUM_GROUPS: 32 10 | WIDTH_PER_GROUP: 8 11 | RPN: 12 | USE_FPN: True 13 | ANCHOR_STRIDE: (4, 8, 16, 32, 64) 14 | PRE_NMS_TOP_N_TRAIN: 2000 15 | PRE_NMS_TOP_N_TEST: 1000 16 | POST_NMS_TOP_N_TEST: 1000 17 | FPN_POST_NMS_TOP_N_TEST: 1000 18 | ROI_HEADS: 19 | USE_FPN: True 20 | ROI_BOX_HEAD: 21 | POOLER_RESOLUTION: 7 22 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 23 | POOLER_SAMPLING_RATIO: 2 24 | FEATURE_EXTRACTOR: "FPN2MLPFeatureExtractor" 25 | PREDICTOR: "FPNPredictor" 26 | ROI_MASK_HEAD: 27 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 28 | FEATURE_EXTRACTOR: "MaskRCNNFPNFeatureExtractor" 29 | PREDICTOR: "MaskRCNNC4Predictor" 30 | POOLER_RESOLUTION: 14 31 | POOLER_SAMPLING_RATIO: 2 32 | RESOLUTION: 28 33 | SHARE_BOX_FEATURE_EXTRACTOR: False 34 | MASK_ON: True 35 | DATASETS: 36 | TEST: ("coco_2014_minival",) 37 | DATALOADER: 38 | SIZE_DIVISIBILITY: 32 39 | -------------------------------------------------------------------------------- /configs/quick_schedules/e2e_faster_rcnn_X_101_32x8d_FPN_quick.yaml: -------------------------------------------------------------------------------- 1 | MODEL: 2 | META_ARCHITECTURE: "GeneralizedRCNN" 3 | WEIGHT: "catalog://ImageNetPretrained/FAIR/20171220/X-101-32x8d" 4 | BACKBONE: 5 | CONV_BODY: "R-101-FPN" 6 | RESNETS: 7 | BACKBONE_OUT_CHANNELS: 256 8 | STRIDE_IN_1X1: False 9 | NUM_GROUPS: 32 10 | WIDTH_PER_GROUP: 8 11 | RPN: 12 | USE_FPN: True 13 | ANCHOR_STRIDE: (4, 8, 16, 32, 64) 14 | PRE_NMS_TOP_N_TRAIN: 2000 15 | PRE_NMS_TOP_N_TEST: 1000 16 | POST_NMS_TOP_N_TEST: 1000 17 | FPN_POST_NMS_TOP_N_TEST: 1000 18 | ROI_HEADS: 19 | USE_FPN: True 20 | BATCH_SIZE_PER_IMAGE: 256 21 | ROI_BOX_HEAD: 22 | POOLER_RESOLUTION: 7 23 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 24 | POOLER_SAMPLING_RATIO: 2 25 | FEATURE_EXTRACTOR: "FPN2MLPFeatureExtractor" 26 | PREDICTOR: "FPNPredictor" 27 | DATASETS: 28 | TRAIN: ("coco_2014_minival",) 29 | TEST: ("coco_2014_minival",) 30 | INPUT: 31 | MIN_SIZE_TRAIN: (600,) 32 | MAX_SIZE_TRAIN: 1000 33 | MIN_SIZE_TEST: 800 34 | MAX_SIZE_TEST: 1000 35 | DATALOADER: 36 | SIZE_DIVISIBILITY: 32 37 | SOLVER: 38 | BASE_LR: 0.005 39 | WEIGHT_DECAY: 0.0001 40 | STEPS: (1500,) 41 | MAX_ITER: 2000 42 | IMS_PER_BATCH: 2 43 | TEST: 44 | IMS_PER_BATCH: 2 45 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/TRW_S-v1.3/treeProbabilities.cpp: -------------------------------------------------------------------------------- 1 | #include 2 | #include 3 | #include 4 | #include 5 | #include 6 | #include "MRFEnergy.h" 7 | 8 | /////////////////////////////////////////////////////////////////////////// 9 | /////////////////////////////////////////////////////////////////////////// 10 | /////////////////////////////////////////////////////////////////////////// 11 | 12 | template void MRFEnergy::SetMonotonicTrees() 13 | { 14 | Node* i; 15 | MRFEdge* e; 16 | 17 | if (!m_isEnergyConstructionCompleted) 18 | { 19 | CompleteGraphConstruction(); 20 | } 21 | 22 | for (i=m_nodeFirst; i; i=i->m_next) 23 | { 24 | REAL mu; 25 | 26 | int nForward = 0, nBackward = 0; 27 | for (e=i->m_firstForward; e; e=e->m_nextForward) 28 | { 29 | nForward ++; 30 | } 31 | for (e=i->m_firstBackward; e; e=e->m_nextBackward) 32 | { 33 | nBackward ++; 34 | } 35 | int ni = (nForward > nBackward) ? nForward : nBackward; 36 | 37 | mu = (REAL)1 / ni; 38 | for (e=i->m_firstBackward; e; e=e->m_nextBackward) 39 | { 40 | e->m_gammaBackward = mu; 41 | } 42 | for (e=i->m_firstForward; e; e=e->m_nextForward) 43 | { 44 | e->m_gammaForward = mu; 45 | } 46 | } 47 | } 48 | 49 | #include "instances.inc" -------------------------------------------------------------------------------- /configs/e2e_mask_rcnn_R_50_FPN_1x.yaml: -------------------------------------------------------------------------------- 1 | MODEL: 2 | META_ARCHITECTURE: "GeneralizedRCNN" 3 | WEIGHT: "catalog://ImageNetPretrained/MSRA/R-50" 4 | BACKBONE: 5 | CONV_BODY: "R-50-FPN" 6 | RESNETS: 7 | BACKBONE_OUT_CHANNELS: 256 8 | RPN: 9 | USE_FPN: True 10 | ANCHOR_STRIDE: (4, 8, 16, 32, 64) 11 | PRE_NMS_TOP_N_TRAIN: 2000 12 | PRE_NMS_TOP_N_TEST: 1000 13 | POST_NMS_TOP_N_TEST: 1000 14 | FPN_POST_NMS_TOP_N_TEST: 1000 15 | ROI_HEADS: 16 | USE_FPN: True 17 | ROI_BOX_HEAD: 18 | POOLER_RESOLUTION: 7 19 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 20 | POOLER_SAMPLING_RATIO: 2 21 | FEATURE_EXTRACTOR: "FPN2MLPFeatureExtractor" 22 | PREDICTOR: "FPNPredictor" 23 | ROI_MASK_HEAD: 24 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 25 | FEATURE_EXTRACTOR: "MaskRCNNFPNFeatureExtractor" 26 | PREDICTOR: "MaskRCNNC4Predictor" 27 | POOLER_RESOLUTION: 14 28 | POOLER_SAMPLING_RATIO: 2 29 | RESOLUTION: 28 30 | SHARE_BOX_FEATURE_EXTRACTOR: False 31 | MASK_ON: True 32 | DATASETS: 33 | TRAIN: ("coco_2014_train", "coco_2014_valminusminival") 34 | TEST: ("coco_2014_minival",) 35 | DATALOADER: 36 | SIZE_DIVISIBILITY: 32 37 | SOLVER: 38 | BASE_LR: 0.02 39 | WEIGHT_DECAY: 0.0001 40 | STEPS: (60000, 80000) 41 | MAX_ITER: 90000 42 | -------------------------------------------------------------------------------- /configs/e2e_mask_rcnn_R_101_FPN_1x.yaml: -------------------------------------------------------------------------------- 1 | MODEL: 2 | META_ARCHITECTURE: "GeneralizedRCNN" 3 | WEIGHT: "catalog://ImageNetPretrained/MSRA/R-101" 4 | BACKBONE: 5 | CONV_BODY: "R-101-FPN" 6 | RESNETS: 7 | BACKBONE_OUT_CHANNELS: 256 8 | RPN: 9 | USE_FPN: True 10 | ANCHOR_STRIDE: (4, 8, 16, 32, 64) 11 | PRE_NMS_TOP_N_TRAIN: 2000 12 | PRE_NMS_TOP_N_TEST: 1000 13 | POST_NMS_TOP_N_TEST: 1000 14 | FPN_POST_NMS_TOP_N_TEST: 1000 15 | ROI_HEADS: 16 | USE_FPN: True 17 | ROI_BOX_HEAD: 18 | POOLER_RESOLUTION: 7 19 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 20 | POOLER_SAMPLING_RATIO: 2 21 | FEATURE_EXTRACTOR: "FPN2MLPFeatureExtractor" 22 | PREDICTOR: "FPNPredictor" 23 | ROI_MASK_HEAD: 24 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 25 | FEATURE_EXTRACTOR: "MaskRCNNFPNFeatureExtractor" 26 | PREDICTOR: "MaskRCNNC4Predictor" 27 | POOLER_RESOLUTION: 14 28 | POOLER_SAMPLING_RATIO: 2 29 | RESOLUTION: 28 30 | SHARE_BOX_FEATURE_EXTRACTOR: False 31 | MASK_ON: True 32 | DATASETS: 33 | TRAIN: ("coco_2014_train", "coco_2014_valminusminival") 34 | TEST: ("coco_2014_minival",) 35 | DATALOADER: 36 | SIZE_DIVISIBILITY: 32 37 | SOLVER: 38 | BASE_LR: 0.02 39 | WEIGHT_DECAY: 0.0001 40 | STEPS: (60000, 80000) 41 | MAX_ITER: 90000 42 | -------------------------------------------------------------------------------- /configs/pascal_voc/e2e_mask_rcnn_R_50_FPN_1x_cocostyle.yaml: -------------------------------------------------------------------------------- 1 | MODEL: 2 | META_ARCHITECTURE: "GeneralizedRCNN" 3 | WEIGHT: "catalog://ImageNetPretrained/MSRA/R-50" 4 | BACKBONE: 5 | CONV_BODY: "R-50-FPN" 6 | RESNETS: 7 | BACKBONE_OUT_CHANNELS: 256 8 | RPN: 9 | USE_FPN: True 10 | ANCHOR_STRIDE: (4, 8, 16, 32, 64) 11 | PRE_NMS_TOP_N_TRAIN: 2000 12 | PRE_NMS_TOP_N_TEST: 1000 13 | POST_NMS_TOP_N_TEST: 1000 14 | FPN_POST_NMS_TOP_N_TEST: 1000 15 | ROI_HEADS: 16 | USE_FPN: True 17 | ROI_BOX_HEAD: 18 | POOLER_RESOLUTION: 7 19 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 20 | POOLER_SAMPLING_RATIO: 2 21 | FEATURE_EXTRACTOR: "FPN2MLPFeatureExtractor" 22 | PREDICTOR: "FPNPredictor" 23 | NUM_CLASSES: 21 24 | ROI_MASK_HEAD: 25 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 26 | FEATURE_EXTRACTOR: "MaskRCNNFPNFeatureExtractor" 27 | PREDICTOR: "MaskRCNNC4Predictor" 28 | POOLER_RESOLUTION: 14 29 | POOLER_SAMPLING_RATIO: 2 30 | RESOLUTION: 28 31 | SHARE_BOX_FEATURE_EXTRACTOR: False 32 | MASK_ON: True 33 | DATASETS: 34 | TRAIN: ("voc_2012_train_cocostyle",) 35 | TEST: ("voc_2012_val_cocostyle",) 36 | DATALOADER: 37 | SIZE_DIVISIBILITY: 32 38 | SOLVER: 39 | BASE_LR: 0.01 40 | WEIGHT_DECAY: 0.0001 41 | STEPS: (18000,) 42 | MAX_ITER: 24000 43 | -------------------------------------------------------------------------------- /maskrcnn_benchmark/data/samplers/iteration_based_batch_sampler.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. 2 | from torch.utils.data.sampler import BatchSampler 3 | 4 | 5 | class IterationBasedBatchSampler(BatchSampler): 6 | """ 7 | Wraps a BatchSampler, resampling from it until 8 | a specified number of iterations have been sampled 9 | """ 10 | 11 | def __init__(self, batch_sampler, num_iterations, start_iter=0): 12 | self.batch_sampler = batch_sampler 13 | self.num_iterations = num_iterations 14 | self.start_iter = start_iter 15 | 16 | def __iter__(self): 17 | iteration = self.start_iter 18 | while iteration <= self.num_iterations: 19 | # if the underlying sampler has a set_epoch method, like 20 | # DistributedSampler, used for making each process see 21 | # a different split of the dataset, then set it 22 | if hasattr(self.batch_sampler.sampler, "set_epoch"): 23 | self.batch_sampler.sampler.set_epoch(iteration) 24 | for batch in self.batch_sampler: 25 | iteration += 1 26 | if iteration > self.num_iterations: 27 | break 28 | yield batch 29 | 30 | def __len__(self): 31 | return self.num_iterations 32 | -------------------------------------------------------------------------------- /maskrcnn_benchmark/utils/timer.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. 2 | 3 | 4 | import time 5 | import datetime 6 | 7 | 8 | class Timer(object): 9 | def __init__(self): 10 | self.reset() 11 | 12 | @property 13 | def average_time(self): 14 | return self.total_time / self.calls if self.calls > 0 else 0.0 15 | 16 | def tic(self): 17 | # using time.time instead of time.clock because time time.clock 18 | # does not normalize for multithreading 19 | self.start_time = time.time() 20 | 21 | def toc(self, average=True): 22 | self.add(time.time() - self.start_time) 23 | if average: 24 | return self.average_time 25 | else: 26 | return self.diff 27 | 28 | def add(self, time_diff): 29 | self.diff = time_diff 30 | self.total_time += self.diff 31 | self.calls += 1 32 | 33 | def reset(self): 34 | self.total_time = 0.0 35 | self.calls = 0 36 | self.start_time = 0.0 37 | self.diff = 0.0 38 | 39 | def avg_time_str(self): 40 | time_str = str(datetime.timedelta(seconds=self.average_time)) 41 | return time_str 42 | 43 | 44 | def get_time_str(time_diff): 45 | time_str = str(datetime.timedelta(seconds=time_diff)) 46 | return time_str 47 | -------------------------------------------------------------------------------- /maskrcnn_benchmark/layers/_utils.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. 2 | import glob 3 | import os.path 4 | 5 | import torch 6 | 7 | try: 8 | from torch.utils.cpp_extension import load as load_ext 9 | from torch.utils.cpp_extension import CUDA_HOME 10 | except ImportError: 11 | raise ImportError("The cpp layer extensions requires PyTorch 0.4 or higher") 12 | 13 | 14 | def _load_C_extensions(): 15 | this_dir = os.path.dirname(os.path.abspath(__file__)) 16 | this_dir = os.path.dirname(this_dir) 17 | this_dir = os.path.join(this_dir, "csrc") 18 | 19 | main_file = glob.glob(os.path.join(this_dir, "*.cpp")) 20 | source_cpu = glob.glob(os.path.join(this_dir, "cpu", "*.cpp")) 21 | source_cuda = glob.glob(os.path.join(this_dir, "cuda", "*.cu")) 22 | 23 | source = main_file + source_cpu 24 | 25 | extra_cflags = [] 26 | if torch.cuda.is_available() and CUDA_HOME is not None: 27 | source.extend(source_cuda) 28 | extra_cflags = ["-DWITH_CUDA"] 29 | source = [os.path.join(this_dir, s) for s in source] 30 | extra_include_paths = [this_dir] 31 | return load_ext( 32 | "torchvision", 33 | source, 34 | extra_cflags=extra_cflags, 35 | extra_include_paths=extra_include_paths, 36 | ) 37 | 38 | 39 | _C = _load_C_extensions() 40 | -------------------------------------------------------------------------------- /configs/cityscapes/e2e_mask_rcnn_R_50_FPN_1x_cocostyle.yaml: -------------------------------------------------------------------------------- 1 | MODEL: 2 | META_ARCHITECTURE: "GeneralizedRCNN" 3 | WEIGHT: "catalog://ImageNetPretrained/MSRA/R-50" 4 | BACKBONE: 5 | CONV_BODY: "R-50-FPN" 6 | RESNETS: 7 | BACKBONE_OUT_CHANNELS: 256 8 | RPN: 9 | USE_FPN: True 10 | ANCHOR_STRIDE: (4, 8, 16, 32, 64) 11 | PRE_NMS_TOP_N_TRAIN: 2000 12 | PRE_NMS_TOP_N_TEST: 1000 13 | POST_NMS_TOP_N_TEST: 1000 14 | FPN_POST_NMS_TOP_N_TEST: 1000 15 | ROI_HEADS: 16 | USE_FPN: True 17 | ROI_BOX_HEAD: 18 | POOLER_RESOLUTION: 7 19 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 20 | POOLER_SAMPLING_RATIO: 2 21 | FEATURE_EXTRACTOR: "FPN2MLPFeatureExtractor" 22 | PREDICTOR: "FPNPredictor" 23 | NUM_CLASSES: 9 24 | ROI_MASK_HEAD: 25 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 26 | FEATURE_EXTRACTOR: "MaskRCNNFPNFeatureExtractor" 27 | PREDICTOR: "MaskRCNNC4Predictor" 28 | POOLER_RESOLUTION: 14 29 | POOLER_SAMPLING_RATIO: 2 30 | RESOLUTION: 28 31 | SHARE_BOX_FEATURE_EXTRACTOR: False 32 | MASK_ON: True 33 | DATASETS: 34 | TRAIN: ("cityscapes_fine_instanceonly_seg_train_cocostyle",) 35 | TEST: ("cityscapes_fine_instanceonly_seg_val_cocostyle",) 36 | DATALOADER: 37 | SIZE_DIVISIBILITY: 32 38 | SOLVER: 39 | BASE_LR: 0.01 40 | WEIGHT_DECAY: 0.0001 41 | STEPS: (18000,) 42 | MAX_ITER: 24000 43 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/UmfPackSupport: -------------------------------------------------------------------------------- 1 | #ifndef EIGEN_UMFPACKSUPPORT_MODULE_H 2 | #define EIGEN_UMFPACKSUPPORT_MODULE_H 3 | 4 | #include "SparseCore" 5 | 6 | #include "src/Core/util/DisableStupidWarnings.h" 7 | 8 | extern "C" { 9 | #include 10 | } 11 | 12 | /** \ingroup Support_modules 13 | * \defgroup UmfPackSupport_Module UmfPackSupport module 14 | * 15 | * This module provides an interface to the UmfPack library which is part of the suitesparse package. 16 | * It provides the following factorization class: 17 | * - class UmfPackLU: a multifrontal sequential LU factorization. 18 | * 19 | * \code 20 | * #include 21 | * \endcode 22 | * 23 | * In order to use this module, the umfpack headers must be accessible from the include paths, and your binary must be linked to the umfpack library and its dependencies. 24 | * The dependencies depend on how umfpack has been compiled. 25 | * For a cmake based project, you can use our FindUmfPack.cmake module to help you in this task. 26 | * 27 | */ 28 | 29 | #include "src/misc/Solve.h" 30 | #include "src/misc/SparseSolve.h" 31 | 32 | #include "src/UmfPackSupport/UmfPackSupport.h" 33 | 34 | #include "src/Core/util/ReenableStupidWarnings.h" 35 | 36 | #endif // EIGEN_UMFPACKSUPPORT_MODULE_H 37 | -------------------------------------------------------------------------------- /configs/retinanet/retinanet_R-50-FPN_1x_quick.yaml: -------------------------------------------------------------------------------- 1 | MODEL: 2 | META_ARCHITECTURE: "GeneralizedRCNN" 3 | WEIGHT: "catalog://ImageNetPretrained/MSRA/R-50" 4 | RPN_ONLY: True 5 | RETINANET_ON: True 6 | BACKBONE: 7 | CONV_BODY: "R-50-FPN-RETINANET" 8 | RESNETS: 9 | BACKBONE_OUT_CHANNELS: 256 10 | RPN: 11 | USE_FPN: True 12 | FG_IOU_THRESHOLD: 0.5 13 | BG_IOU_THRESHOLD: 0.4 14 | ANCHOR_STRIDE: (4, 8, 16, 32, 64) 15 | PRE_NMS_TOP_N_TRAIN: 2000 16 | PRE_NMS_TOP_N_TEST: 1000 17 | POST_NMS_TOP_N_TEST: 1000 18 | FPN_POST_NMS_TOP_N_TEST: 1000 19 | ROI_HEADS: 20 | USE_FPN: True 21 | BATCH_SIZE_PER_IMAGE: 256 22 | ROI_BOX_HEAD: 23 | POOLER_RESOLUTION: 7 24 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 25 | POOLER_SAMPLING_RATIO: 2 26 | FEATURE_EXTRACTOR: "FPN2MLPFeatureExtractor" 27 | PREDICTOR: "FPNPredictor" 28 | RETINANET: 29 | SCALES_PER_OCTAVE: 3 30 | STRADDLE_THRESH: -1 31 | FG_IOU_THRESHOLD: 0.5 32 | BG_IOU_THRESHOLD: 0.4 33 | DATASETS: 34 | TRAIN: ("coco_2014_minival",) 35 | TEST: ("coco_2014_minival",) 36 | INPUT: 37 | MIN_SIZE_TRAIN: (600,) 38 | MAX_SIZE_TRAIN: 1000 39 | MIN_SIZE_TEST: 800 40 | MAX_SIZE_TEST: 1000 41 | DATALOADER: 42 | SIZE_DIVISIBILITY: 32 43 | SOLVER: 44 | BASE_LR: 0.005 45 | WEIGHT_DECAY: 0.0001 46 | STEPS: (3500,) 47 | MAX_ITER: 4000 48 | IMS_PER_BATCH: 4 49 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/compile.m: -------------------------------------------------------------------------------- 1 | % Compile files only if needed 2 | function compile(cpp_file, out_file, sources, extra_arguments) 3 | 4 | % Process 5 | my_name = mfilename('fullpath'); 6 | my_path = [fileparts(my_name)]; 7 | mex_file_name = [my_path filesep out_file '.' mexext]; 8 | cpp_file_name = [my_path filesep cpp_file]; 9 | 10 | mex_file = dir(mex_file_name); 11 | cpp_file = dir(cpp_file_name); 12 | 13 | if length(mex_file) == 1 14 | mex_modified = mex_file.datenum; 15 | else 16 | mex_modified = 0; 17 | end 18 | 19 | cpp_modified = cpp_file.datenum; 20 | 21 | % If modified ocd r not existant compile 22 | compile_file = false; 23 | if ~exist(mex_file_name, 'file') 24 | compile_file = true; 25 | elseif mex_modified < cpp_modified 26 | compile_file = true; 27 | end 28 | 29 | include_folders = {}; 30 | 31 | % Append current folder to sources 32 | for i = 1 : length(sources); 33 | include_folders{i} = ['-I' my_path filesep fileparts(sources{i}) filesep]; 34 | sources{i} = [my_path filesep sources{i}]; 35 | end 36 | 37 | % Check all dependend files 38 | for i = 1 : length(sources) 39 | cpp_file = dir(sources{i}); 40 | cpp_modified = cpp_file.datenum; 41 | if mex_modified < cpp_modified 42 | compile_file = true; 43 | end 44 | end 45 | 46 | args = {extra_arguments{:},include_folders{:},sources{:}}; 47 | 48 | if compile_file 49 | mex(cpp_file_name,'-outdir',my_path, args{:}); 50 | end 51 | -------------------------------------------------------------------------------- /maskrcnn_benchmark/utils/env.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. 2 | import os 3 | 4 | from maskrcnn_benchmark.utils.imports import import_file 5 | 6 | 7 | def setup_environment(): 8 | """Perform environment setup work. The default setup is a no-op, but this 9 | function allows the user to specify a Python source file that performs 10 | custom setup work that may be necessary to their computing environment. 11 | """ 12 | custom_module_path = os.environ.get("TORCH_DETECTRON_ENV_MODULE") 13 | if custom_module_path: 14 | setup_custom_environment(custom_module_path) 15 | else: 16 | # The default setup is a no-op 17 | pass 18 | 19 | 20 | def setup_custom_environment(custom_module_path): 21 | """Load custom environment setup from a Python source file and run the setup 22 | function. 23 | """ 24 | module = import_file("maskrcnn_benchmark.utils.env.custom_module", custom_module_path) 25 | assert hasattr(module, "setup_environment") and callable( 26 | module.setup_environment 27 | ), ( 28 | "Custom environment module defined in {} does not have the " 29 | "required callable attribute 'setup_environment'." 30 | ).format( 31 | custom_module_path 32 | ) 33 | module.setup_environment() 34 | 35 | 36 | # Force environment setup when this module is imported 37 | setup_environment() 38 | -------------------------------------------------------------------------------- /configs/e2e_keypoint_rcnn_R_50_FPN_1x.yaml: -------------------------------------------------------------------------------- 1 | MODEL: 2 | META_ARCHITECTURE: "GeneralizedRCNN" 3 | WEIGHT: "catalog://ImageNetPretrained/MSRA/R-50" 4 | BACKBONE: 5 | CONV_BODY: "R-50-FPN" 6 | RESNETS: 7 | BACKBONE_OUT_CHANNELS: 256 8 | RPN: 9 | USE_FPN: True 10 | ANCHOR_STRIDE: (4, 8, 16, 32, 64) 11 | PRE_NMS_TOP_N_TRAIN: 2000 12 | PRE_NMS_TOP_N_TEST: 1000 13 | POST_NMS_TOP_N_TEST: 1000 14 | FPN_POST_NMS_TOP_N_TEST: 1000 15 | ROI_HEADS: 16 | USE_FPN: True 17 | ROI_BOX_HEAD: 18 | POOLER_RESOLUTION: 7 19 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 20 | POOLER_SAMPLING_RATIO: 2 21 | FEATURE_EXTRACTOR: "FPN2MLPFeatureExtractor" 22 | PREDICTOR: "FPNPredictor" 23 | NUM_CLASSES: 2 24 | ROI_KEYPOINT_HEAD: 25 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 26 | FEATURE_EXTRACTOR: "KeypointRCNNFeatureExtractor" 27 | PREDICTOR: "KeypointRCNNPredictor" 28 | POOLER_RESOLUTION: 14 29 | POOLER_SAMPLING_RATIO: 2 30 | RESOLUTION: 56 31 | SHARE_BOX_FEATURE_EXTRACTOR: False 32 | KEYPOINT_ON: True 33 | DATASETS: 34 | TRAIN: ("keypoints_coco_2014_train", "keypoints_coco_2014_valminusminival",) 35 | TEST: ("keypoints_coco_2014_minival",) 36 | INPUT: 37 | MIN_SIZE_TRAIN: (640, 672, 704, 736, 768, 800) 38 | DATALOADER: 39 | SIZE_DIVISIBILITY: 32 40 | SOLVER: 41 | BASE_LR: 0.02 42 | WEIGHT_DECAY: 0.0001 43 | STEPS: (60000, 80000) 44 | MAX_ITER: 90000 45 | -------------------------------------------------------------------------------- /configs/e2e_mask_rcnn_X_101_32x8d_FPN_1x.yaml: -------------------------------------------------------------------------------- 1 | MODEL: 2 | META_ARCHITECTURE: "GeneralizedRCNN" 3 | WEIGHT: "catalog://ImageNetPretrained/FAIR/20171220/X-101-32x8d" 4 | BACKBONE: 5 | CONV_BODY: "R-101-FPN" 6 | RESNETS: 7 | BACKBONE_OUT_CHANNELS: 256 8 | STRIDE_IN_1X1: False 9 | NUM_GROUPS: 32 10 | WIDTH_PER_GROUP: 8 11 | RPN: 12 | USE_FPN: True 13 | ANCHOR_STRIDE: (4, 8, 16, 32, 64) 14 | PRE_NMS_TOP_N_TRAIN: 2000 15 | PRE_NMS_TOP_N_TEST: 1000 16 | POST_NMS_TOP_N_TEST: 1000 17 | FPN_POST_NMS_TOP_N_TEST: 1000 18 | ROI_HEADS: 19 | USE_FPN: True 20 | ROI_BOX_HEAD: 21 | POOLER_RESOLUTION: 7 22 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 23 | POOLER_SAMPLING_RATIO: 2 24 | FEATURE_EXTRACTOR: "FPN2MLPFeatureExtractor" 25 | PREDICTOR: "FPNPredictor" 26 | ROI_MASK_HEAD: 27 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 28 | FEATURE_EXTRACTOR: "MaskRCNNFPNFeatureExtractor" 29 | PREDICTOR: "MaskRCNNC4Predictor" 30 | POOLER_RESOLUTION: 14 31 | POOLER_SAMPLING_RATIO: 2 32 | RESOLUTION: 28 33 | SHARE_BOX_FEATURE_EXTRACTOR: False 34 | MASK_ON: True 35 | DATASETS: 36 | TRAIN: ("coco_2014_train", "coco_2014_valminusminival") 37 | TEST: ("coco_2014_minival",) 38 | DATALOADER: 39 | SIZE_DIVISIBILITY: 32 40 | SOLVER: 41 | BASE_LR: 0.01 42 | WEIGHT_DECAY: 0.0001 43 | STEPS: (120000, 160000) 44 | MAX_ITER: 180000 45 | IMS_PER_BATCH: 8 46 | -------------------------------------------------------------------------------- /configs/retinanet/retinanet_R-50-FPN_1x.yaml: -------------------------------------------------------------------------------- 1 | MODEL: 2 | META_ARCHITECTURE: "GeneralizedRCNN" 3 | WEIGHT: "catalog://ImageNetPretrained/MSRA/R-50" 4 | RPN_ONLY: True 5 | RETINANET_ON: True 6 | BACKBONE: 7 | CONV_BODY: "R-50-FPN-RETINANET" 8 | RESNETS: 9 | BACKBONE_OUT_CHANNELS: 256 10 | RPN: 11 | USE_FPN: True 12 | FG_IOU_THRESHOLD: 0.5 13 | BG_IOU_THRESHOLD: 0.4 14 | ANCHOR_STRIDE: (4, 8, 16, 32, 64) 15 | PRE_NMS_TOP_N_TRAIN: 2000 16 | PRE_NMS_TOP_N_TEST: 1000 17 | POST_NMS_TOP_N_TEST: 1000 18 | FPN_POST_NMS_TOP_N_TEST: 1000 19 | ROI_HEADS: 20 | USE_FPN: True 21 | BATCH_SIZE_PER_IMAGE: 256 22 | ROI_BOX_HEAD: 23 | POOLER_RESOLUTION: 7 24 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 25 | POOLER_SAMPLING_RATIO: 2 26 | FEATURE_EXTRACTOR: "FPN2MLPFeatureExtractor" 27 | PREDICTOR: "FPNPredictor" 28 | RETINANET: 29 | SCALES_PER_OCTAVE: 3 30 | STRADDLE_THRESH: -1 31 | FG_IOU_THRESHOLD: 0.5 32 | BG_IOU_THRESHOLD: 0.4 33 | DATASETS: 34 | TRAIN: ("coco_2014_train", "coco_2014_valminusminival") 35 | TEST: ("coco_2014_minival",) 36 | INPUT: 37 | MIN_SIZE_TRAIN: (800,) 38 | MAX_SIZE_TRAIN: 1333 39 | MIN_SIZE_TEST: 800 40 | MAX_SIZE_TEST: 1333 41 | DATALOADER: 42 | SIZE_DIVISIBILITY: 32 43 | SOLVER: 44 | # Assume 4 gpus 45 | BASE_LR: 0.005 46 | WEIGHT_DECAY: 0.0001 47 | STEPS: (120000, 160000) 48 | MAX_ITER: 180000 49 | IMS_PER_BATCH: 8 50 | -------------------------------------------------------------------------------- /configs/retinanet/retinanet_R-101-FPN_1x.yaml: -------------------------------------------------------------------------------- 1 | MODEL: 2 | META_ARCHITECTURE: "GeneralizedRCNN" 3 | WEIGHT: "catalog://ImageNetPretrained/MSRA/R-101" 4 | RPN_ONLY: True 5 | RETINANET_ON: True 6 | BACKBONE: 7 | CONV_BODY: "R-101-FPN-RETINANET" 8 | RESNETS: 9 | BACKBONE_OUT_CHANNELS: 256 10 | RPN: 11 | USE_FPN: True 12 | FG_IOU_THRESHOLD: 0.5 13 | BG_IOU_THRESHOLD: 0.4 14 | ANCHOR_STRIDE: (4, 8, 16, 32, 64) 15 | PRE_NMS_TOP_N_TRAIN: 2000 16 | PRE_NMS_TOP_N_TEST: 1000 17 | POST_NMS_TOP_N_TEST: 1000 18 | FPN_POST_NMS_TOP_N_TEST: 1000 19 | ROI_HEADS: 20 | USE_FPN: True 21 | BATCH_SIZE_PER_IMAGE: 256 22 | ROI_BOX_HEAD: 23 | POOLER_RESOLUTION: 7 24 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 25 | POOLER_SAMPLING_RATIO: 2 26 | FEATURE_EXTRACTOR: "FPN2MLPFeatureExtractor" 27 | PREDICTOR: "FPNPredictor" 28 | RETINANET: 29 | SCALES_PER_OCTAVE: 3 30 | STRADDLE_THRESH: -1 31 | FG_IOU_THRESHOLD: 0.5 32 | BG_IOU_THRESHOLD: 0.4 33 | DATASETS: 34 | TRAIN: ("coco_2014_train", "coco_2014_valminusminival") 35 | TEST: ("coco_2014_minival",) 36 | INPUT: 37 | MIN_SIZE_TRAIN: (800, ) 38 | MAX_SIZE_TRAIN: 1333 39 | MIN_SIZE_TEST: 800 40 | MAX_SIZE_TEST: 1333 41 | DATALOADER: 42 | SIZE_DIVISIBILITY: 32 43 | SOLVER: 44 | # Assume 4 gpus 45 | BASE_LR: 0.005 46 | WEIGHT_DECAY: 0.0001 47 | STEPS: (120000, 160000) 48 | MAX_ITER: 180000 49 | IMS_PER_BATCH: 8 50 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/external/liblbfgs/lbfgs.sln: -------------------------------------------------------------------------------- 1 | 2 | Microsoft Visual Studio Solution File, Format Version 11.00 3 | # Visual Studio 2010 4 | Project("{8BC9CEB8-8B4A-11D0-8D11-00A0C91BC942}") = "lib", "lib\lib.vcxproj", "{C4405B73-A899-44BF-8681-04CE040B6705}" 5 | EndProject 6 | Project("{8BC9CEB8-8B4A-11D0-8D11-00A0C91BC942}") = "sample", "sample\sample.vcxproj", "{B4D7D5F5-4A4E-49D5-B38A-E5673520DE66}" 7 | EndProject 8 | Global 9 | GlobalSection(SolutionConfigurationPlatforms) = preSolution 10 | Debug|Win32 = Debug|Win32 11 | Release|Win32 = Release|Win32 12 | EndGlobalSection 13 | GlobalSection(ProjectConfigurationPlatforms) = postSolution 14 | {C4405B73-A899-44BF-8681-04CE040B6705}.Debug|Win32.ActiveCfg = Debug|Win32 15 | {C4405B73-A899-44BF-8681-04CE040B6705}.Debug|Win32.Build.0 = Debug|Win32 16 | {C4405B73-A899-44BF-8681-04CE040B6705}.Release|Win32.ActiveCfg = Release|Win32 17 | {C4405B73-A899-44BF-8681-04CE040B6705}.Release|Win32.Build.0 = Release|Win32 18 | {B4D7D5F5-4A4E-49D5-B38A-E5673520DE66}.Debug|Win32.ActiveCfg = Debug|Win32 19 | {B4D7D5F5-4A4E-49D5-B38A-E5673520DE66}.Debug|Win32.Build.0 = Debug|Win32 20 | {B4D7D5F5-4A4E-49D5-B38A-E5673520DE66}.Release|Win32.ActiveCfg = Release|Win32 21 | {B4D7D5F5-4A4E-49D5-B38A-E5673520DE66}.Release|Win32.Build.0 = Release|Win32 22 | EndGlobalSection 23 | GlobalSection(SolutionProperties) = preSolution 24 | HideSolutionNode = FALSE 25 | EndGlobalSection 26 | EndGlobal 27 | -------------------------------------------------------------------------------- /configs/retinanet/retinanet_R-50-FPN_P5_1x.yaml: -------------------------------------------------------------------------------- 1 | MODEL: 2 | META_ARCHITECTURE: "GeneralizedRCNN" 3 | WEIGHT: "catalog://ImageNetPretrained/MSRA/R-50" 4 | RPN_ONLY: True 5 | RETINANET_ON: True 6 | BACKBONE: 7 | CONV_BODY: "R-50-FPN-RETINANET" 8 | RESNETS: 9 | BACKBONE_OUT_CHANNELS: 256 10 | RPN: 11 | USE_FPN: True 12 | FG_IOU_THRESHOLD: 0.5 13 | BG_IOU_THRESHOLD: 0.4 14 | ANCHOR_STRIDE: (4, 8, 16, 32, 64) 15 | PRE_NMS_TOP_N_TRAIN: 2000 16 | PRE_NMS_TOP_N_TEST: 1000 17 | POST_NMS_TOP_N_TEST: 1000 18 | FPN_POST_NMS_TOP_N_TEST: 1000 19 | ROI_HEADS: 20 | USE_FPN: True 21 | BATCH_SIZE_PER_IMAGE: 256 22 | ROI_BOX_HEAD: 23 | POOLER_RESOLUTION: 7 24 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 25 | POOLER_SAMPLING_RATIO: 2 26 | FEATURE_EXTRACTOR: "FPN2MLPFeatureExtractor" 27 | PREDICTOR: "FPNPredictor" 28 | RETINANET: 29 | SCALES_PER_OCTAVE: 3 30 | STRADDLE_THRESH: -1 31 | USE_C5: False 32 | FG_IOU_THRESHOLD: 0.5 33 | BG_IOU_THRESHOLD: 0.4 34 | DATASETS: 35 | TRAIN: ("coco_2014_train", "coco_2014_valminusminival") 36 | TEST: ("coco_2014_minival",) 37 | INPUT: 38 | MIN_SIZE_TRAIN: (800,) 39 | MAX_SIZE_TRAIN: 1333 40 | MIN_SIZE_TEST: 800 41 | MAX_SIZE_TEST: 1333 42 | DATALOADER: 43 | SIZE_DIVISIBILITY: 32 44 | SOLVER: 45 | # Assume 4 gpus 46 | BASE_LR: 0.005 47 | WEIGHT_DECAY: 0.0001 48 | STEPS: (120000, 160000) 49 | MAX_ITER: 180000 50 | IMS_PER_BATCH: 8 51 | -------------------------------------------------------------------------------- /configs/retinanet/retinanet_R-101-FPN_P5_1x.yaml: -------------------------------------------------------------------------------- 1 | MODEL: 2 | META_ARCHITECTURE: "GeneralizedRCNN" 3 | WEIGHT: "catalog://ImageNetPretrained/MSRA/R-101" 4 | RPN_ONLY: True 5 | RETINANET_ON: True 6 | BACKBONE: 7 | CONV_BODY: "R-101-FPN-RETINANET" 8 | RESNETS: 9 | BACKBONE_OUT_CHANNELS: 256 10 | RPN: 11 | USE_FPN: True 12 | FG_IOU_THRESHOLD: 0.5 13 | BG_IOU_THRESHOLD: 0.4 14 | ANCHOR_STRIDE: (4, 8, 16, 32, 64) 15 | PRE_NMS_TOP_N_TRAIN: 2000 16 | PRE_NMS_TOP_N_TEST: 1000 17 | POST_NMS_TOP_N_TEST: 1000 18 | FPN_POST_NMS_TOP_N_TEST: 1000 19 | ROI_HEADS: 20 | USE_FPN: True 21 | BATCH_SIZE_PER_IMAGE: 256 22 | ROI_BOX_HEAD: 23 | POOLER_RESOLUTION: 7 24 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 25 | POOLER_SAMPLING_RATIO: 2 26 | FEATURE_EXTRACTOR: "FPN2MLPFeatureExtractor" 27 | PREDICTOR: "FPNPredictor" 28 | RETINANET: 29 | SCALES_PER_OCTAVE: 3 30 | STRADDLE_THRESH: -1 31 | USE_C5: False 32 | FG_IOU_THRESHOLD: 0.5 33 | BG_IOU_THRESHOLD: 0.4 34 | DATASETS: 35 | TRAIN: ("coco_2014_train", "coco_2014_valminusminival") 36 | TEST: ("coco_2014_minival",) 37 | INPUT: 38 | MIN_SIZE_TRAIN: (800, ) 39 | MAX_SIZE_TRAIN: 1333 40 | MIN_SIZE_TEST: 800 41 | MAX_SIZE_TEST: 1333 42 | DATALOADER: 43 | SIZE_DIVISIBILITY: 32 44 | SOLVER: 45 | # Assume 4 gpus 46 | BASE_LR: 0.005 47 | WEIGHT_DECAY: 0.0001 48 | STEPS: (120000, 160000) 49 | MAX_ITER: 180000 50 | IMS_PER_BATCH: 8 51 | -------------------------------------------------------------------------------- /configs/caffe2/e2e_keypoint_rcnn_R_50_FPN_1x_caffe2.yaml: -------------------------------------------------------------------------------- 1 | MODEL: 2 | META_ARCHITECTURE: "GeneralizedRCNN" 3 | WEIGHT: "catalog://Caffe2Detectron/COCO/37697547/e2e_keypoint_rcnn_R-50-FPN_1x" 4 | BACKBONE: 5 | CONV_BODY: "R-50-FPN" 6 | RESNETS: 7 | BACKBONE_OUT_CHANNELS: 256 8 | RPN: 9 | USE_FPN: True 10 | ANCHOR_STRIDE: (4, 8, 16, 32, 64) 11 | PRE_NMS_TOP_N_TRAIN: 2000 12 | PRE_NMS_TOP_N_TEST: 1000 13 | POST_NMS_TOP_N_TEST: 1000 14 | FPN_POST_NMS_TOP_N_TEST: 1000 15 | ROI_HEADS: 16 | USE_FPN: True 17 | ROI_BOX_HEAD: 18 | POOLER_RESOLUTION: 7 19 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 20 | POOLER_SAMPLING_RATIO: 2 21 | FEATURE_EXTRACTOR: "FPN2MLPFeatureExtractor" 22 | PREDICTOR: "FPNPredictor" 23 | NUM_CLASSES: 2 24 | ROI_KEYPOINT_HEAD: 25 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 26 | FEATURE_EXTRACTOR: "KeypointRCNNFeatureExtractor" 27 | PREDICTOR: "KeypointRCNNPredictor" 28 | POOLER_RESOLUTION: 14 29 | POOLER_SAMPLING_RATIO: 2 30 | RESOLUTION: 56 31 | SHARE_BOX_FEATURE_EXTRACTOR: False 32 | KEYPOINT_ON: True 33 | DATASETS: 34 | TRAIN: ("keypoints_coco_2014_train", "keypoints_coco_2014_valminusminival",) 35 | TEST: ("keypoints_coco_2014_minival",) 36 | INPUT: 37 | MIN_SIZE_TRAIN: (640, 672, 704, 736, 768, 800) 38 | DATALOADER: 39 | SIZE_DIVISIBILITY: 32 40 | SOLVER: 41 | BASE_LR: 0.02 42 | WEIGHT_DECAY: 0.0001 43 | STEPS: (60000, 80000) 44 | MAX_ITER: 90000 45 | -------------------------------------------------------------------------------- /configs/gn_baselines/e2e_faster_rcnn_R_50_FPN_1x_gn.yaml: -------------------------------------------------------------------------------- 1 | INPUT: 2 | MIN_SIZE_TRAIN: (800,) 3 | MAX_SIZE_TRAIN: 1333 4 | MIN_SIZE_TEST: 800 5 | MAX_SIZE_TEST: 1333 6 | MODEL: 7 | META_ARCHITECTURE: "GeneralizedRCNN" 8 | WEIGHT: "catalog://ImageNetPretrained/MSRA/R-50-GN" 9 | BACKBONE: 10 | CONV_BODY: "R-50-FPN" 11 | RESNETS: # use GN for backbone 12 | BACKBONE_OUT_CHANNELS: 256 13 | STRIDE_IN_1X1: False 14 | TRANS_FUNC: "BottleneckWithGN" 15 | STEM_FUNC: "StemWithGN" 16 | FPN: 17 | USE_GN: True # use GN for FPN 18 | RPN: 19 | USE_FPN: True 20 | ANCHOR_STRIDE: (4, 8, 16, 32, 64) 21 | PRE_NMS_TOP_N_TRAIN: 2000 22 | PRE_NMS_TOP_N_TEST: 1000 23 | POST_NMS_TOP_N_TEST: 1000 24 | FPN_POST_NMS_TOP_N_TEST: 1000 25 | ROI_HEADS: 26 | USE_FPN: True 27 | BATCH_SIZE_PER_IMAGE: 512 28 | POSITIVE_FRACTION: 0.25 29 | ROI_BOX_HEAD: 30 | USE_GN: True # use GN for bbox head 31 | POOLER_RESOLUTION: 7 32 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 33 | POOLER_SAMPLING_RATIO: 2 34 | FEATURE_EXTRACTOR: "FPN2MLPFeatureExtractor" 35 | PREDICTOR: "FPNPredictor" 36 | DATASETS: 37 | TRAIN: ("coco_2014_train", "coco_2014_valminusminival") 38 | TEST: ("coco_2014_minival",) 39 | DATALOADER: 40 | SIZE_DIVISIBILITY: 32 41 | SOLVER: 42 | # Assume 8 gpus 43 | BASE_LR: 0.02 44 | WEIGHT_DECAY: 0.0001 45 | STEPS: (60000, 80000) 46 | MAX_ITER: 90000 47 | IMS_PER_BATCH: 16 48 | TEST: 49 | IMS_PER_BATCH: 8 50 | -------------------------------------------------------------------------------- /maskrcnn_benchmark/modeling/roi_heads/keypoint_head/roi_keypoint_predictors.py: -------------------------------------------------------------------------------- 1 | from torch import nn 2 | 3 | from maskrcnn_benchmark import layers 4 | from maskrcnn_benchmark.modeling import registry 5 | 6 | 7 | @registry.ROI_KEYPOINT_PREDICTOR.register("KeypointRCNNPredictor") 8 | class KeypointRCNNPredictor(nn.Module): 9 | def __init__(self, cfg, in_channels): 10 | super(KeypointRCNNPredictor, self).__init__() 11 | input_features = in_channels 12 | num_keypoints = cfg.MODEL.ROI_KEYPOINT_HEAD.NUM_CLASSES 13 | deconv_kernel = 4 14 | self.kps_score_lowres = layers.ConvTranspose2d( 15 | input_features, 16 | num_keypoints, 17 | deconv_kernel, 18 | stride=2, 19 | padding=deconv_kernel // 2 - 1, 20 | ) 21 | nn.init.kaiming_normal_( 22 | self.kps_score_lowres.weight, mode="fan_out", nonlinearity="relu" 23 | ) 24 | nn.init.constant_(self.kps_score_lowres.bias, 0) 25 | self.up_scale = 2 26 | self.out_channels = num_keypoints 27 | 28 | def forward(self, x): 29 | x = self.kps_score_lowres(x) 30 | x = layers.interpolate( 31 | x, scale_factor=self.up_scale, mode="bilinear", align_corners=False 32 | ) 33 | return x 34 | 35 | 36 | def make_roi_keypoint_predictor(cfg, in_channels): 37 | func = registry.ROI_KEYPOINT_PREDICTOR[cfg.MODEL.ROI_KEYPOINT_HEAD.PREDICTOR] 38 | return func(cfg, in_channels) 39 | -------------------------------------------------------------------------------- /configs/quick_schedules/e2e_mask_rcnn_R_50_FPN_quick.yaml: -------------------------------------------------------------------------------- 1 | MODEL: 2 | META_ARCHITECTURE: "GeneralizedRCNN" 3 | WEIGHT: "catalog://ImageNetPretrained/MSRA/R-50" 4 | BACKBONE: 5 | CONV_BODY: "R-50-FPN" 6 | RESNETS: 7 | BACKBONE_OUT_CHANNELS: 256 8 | RPN: 9 | USE_FPN: True 10 | ANCHOR_STRIDE: (4, 8, 16, 32, 64) 11 | PRE_NMS_TOP_N_TRAIN: 2000 12 | PRE_NMS_TOP_N_TEST: 1000 13 | POST_NMS_TOP_N_TEST: 1000 14 | FPN_POST_NMS_TOP_N_TEST: 1000 15 | ROI_HEADS: 16 | USE_FPN: True 17 | BATCH_SIZE_PER_IMAGE: 256 18 | ROI_BOX_HEAD: 19 | POOLER_RESOLUTION: 7 20 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 21 | POOLER_SAMPLING_RATIO: 2 22 | FEATURE_EXTRACTOR: "FPN2MLPFeatureExtractor" 23 | PREDICTOR: "FPNPredictor" 24 | ROI_MASK_HEAD: 25 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 26 | FEATURE_EXTRACTOR: "MaskRCNNFPNFeatureExtractor" 27 | PREDICTOR: "MaskRCNNC4Predictor" 28 | POOLER_RESOLUTION: 14 29 | POOLER_SAMPLING_RATIO: 2 30 | RESOLUTION: 28 31 | SHARE_BOX_FEATURE_EXTRACTOR: False 32 | MASK_ON: True 33 | DATASETS: 34 | TRAIN: ("coco_2014_minival",) 35 | TEST: ("coco_2014_minival",) 36 | INPUT: 37 | MIN_SIZE_TRAIN: (600,) 38 | MAX_SIZE_TRAIN: 1000 39 | MIN_SIZE_TEST: 800 40 | MAX_SIZE_TEST: 1000 41 | DATALOADER: 42 | SIZE_DIVISIBILITY: 32 43 | SOLVER: 44 | BASE_LR: 0.005 45 | WEIGHT_DECAY: 0.0001 46 | STEPS: (1500,) 47 | MAX_ITER: 2000 48 | IMS_PER_BATCH: 4 49 | TEST: 50 | IMS_PER_BATCH: 2 51 | -------------------------------------------------------------------------------- /configs/e2e_mask_rcnn_fbnet.yaml: -------------------------------------------------------------------------------- 1 | MODEL: 2 | META_ARCHITECTURE: "GeneralizedRCNN" 3 | BACKBONE: 4 | CONV_BODY: FBNet 5 | FBNET: 6 | ARCH: "default" 7 | BN_TYPE: "bn" 8 | WIDTH_DIVISOR: 8 9 | DW_CONV_SKIP_BN: True 10 | DW_CONV_SKIP_RELU: True 11 | DET_HEAD_LAST_SCALE: 0.0 12 | RPN: 13 | ANCHOR_SIZES: (16, 32, 64, 128, 256) 14 | ANCHOR_STRIDE: (16, ) 15 | BATCH_SIZE_PER_IMAGE: 256 16 | PRE_NMS_TOP_N_TRAIN: 6000 17 | PRE_NMS_TOP_N_TEST: 6000 18 | POST_NMS_TOP_N_TRAIN: 2000 19 | POST_NMS_TOP_N_TEST: 100 20 | RPN_HEAD: FBNet.rpn_head 21 | ROI_HEADS: 22 | BATCH_SIZE_PER_IMAGE: 256 23 | ROI_BOX_HEAD: 24 | POOLER_RESOLUTION: 6 25 | FEATURE_EXTRACTOR: FBNet.roi_head 26 | NUM_CLASSES: 81 27 | ROI_MASK_HEAD: 28 | POOLER_RESOLUTION: 6 29 | FEATURE_EXTRACTOR: FBNet.roi_head_mask 30 | PREDICTOR: "MaskRCNNConv1x1Predictor" 31 | RESOLUTION: 12 32 | SHARE_BOX_FEATURE_EXTRACTOR: False 33 | MASK_ON: True 34 | DATASETS: 35 | TRAIN: ("coco_2014_train", "coco_2014_valminusminival") 36 | TEST: ("coco_2014_minival",) 37 | SOLVER: 38 | BASE_LR: 0.06 39 | WARMUP_FACTOR: 0.1 40 | WEIGHT_DECAY: 0.0001 41 | STEPS: (60000, 80000) 42 | MAX_ITER: 90000 43 | IMS_PER_BATCH: 128 # for 8GPUs 44 | # TEST: 45 | # IMS_PER_BATCH: 8 46 | INPUT: 47 | MIN_SIZE_TRAIN: (320, ) 48 | MAX_SIZE_TRAIN: 640 49 | MIN_SIZE_TEST: 320 50 | MAX_SIZE_TEST: 640 51 | PIXEL_MEAN: [103.53, 116.28, 123.675] 52 | PIXEL_STD: [57.375, 57.12, 58.395] 53 | -------------------------------------------------------------------------------- /configs/e2e_mask_rcnn_fbnet_600.yaml: -------------------------------------------------------------------------------- 1 | MODEL: 2 | META_ARCHITECTURE: "GeneralizedRCNN" 3 | BACKBONE: 4 | CONV_BODY: FBNet 5 | FBNET: 6 | ARCH: "default" 7 | BN_TYPE: "bn" 8 | WIDTH_DIVISOR: 8 9 | DW_CONV_SKIP_BN: True 10 | DW_CONV_SKIP_RELU: True 11 | DET_HEAD_LAST_SCALE: 0.0 12 | RPN: 13 | ANCHOR_SIZES: (32, 64, 128, 256, 512) 14 | ANCHOR_STRIDE: (16, ) 15 | BATCH_SIZE_PER_IMAGE: 256 16 | PRE_NMS_TOP_N_TRAIN: 6000 17 | PRE_NMS_TOP_N_TEST: 6000 18 | POST_NMS_TOP_N_TRAIN: 2000 19 | POST_NMS_TOP_N_TEST: 200 20 | RPN_HEAD: FBNet.rpn_head 21 | ROI_HEADS: 22 | BATCH_SIZE_PER_IMAGE: 256 23 | ROI_BOX_HEAD: 24 | POOLER_RESOLUTION: 6 25 | FEATURE_EXTRACTOR: FBNet.roi_head 26 | NUM_CLASSES: 81 27 | ROI_MASK_HEAD: 28 | POOLER_RESOLUTION: 6 29 | FEATURE_EXTRACTOR: FBNet.roi_head_mask 30 | PREDICTOR: "MaskRCNNConv1x1Predictor" 31 | RESOLUTION: 12 32 | SHARE_BOX_FEATURE_EXTRACTOR: False 33 | MASK_ON: True 34 | DATASETS: 35 | TRAIN: ("coco_2014_train", "coco_2014_valminusminival") 36 | TEST: ("coco_2014_minival",) 37 | SOLVER: 38 | BASE_LR: 0.06 39 | WARMUP_FACTOR: 0.1 40 | WEIGHT_DECAY: 0.0001 41 | STEPS: (60000, 80000) 42 | MAX_ITER: 90000 43 | IMS_PER_BATCH: 128 # for 8GPUs 44 | # TEST: 45 | # IMS_PER_BATCH: 8 46 | INPUT: 47 | MIN_SIZE_TRAIN: (600, ) 48 | MAX_SIZE_TRAIN: 1000 49 | MIN_SIZE_TEST: 600 50 | MAX_SIZE_TEST: 1000 51 | PIXEL_MEAN: [103.53, 116.28, 123.675] 52 | PIXEL_STD: [57.375, 57.12, 58.395] 53 | -------------------------------------------------------------------------------- /configs/gn_baselines/scratch_e2e_faster_rcnn_R_50_FPN_3x_gn.yaml: -------------------------------------------------------------------------------- 1 | INPUT: 2 | MIN_SIZE_TRAIN: (800,) 3 | MAX_SIZE_TRAIN: 1333 4 | MIN_SIZE_TEST: 800 5 | MAX_SIZE_TEST: 1333 6 | MODEL: 7 | META_ARCHITECTURE: "GeneralizedRCNN" 8 | WEIGHT: "" # no pretrained model 9 | BACKBONE: 10 | CONV_BODY: "R-50-FPN" 11 | FREEZE_CONV_BODY_AT: 0 # finetune all layers 12 | RESNETS: # use GN for backbone 13 | BACKBONE_OUT_CHANNELS: 256 14 | STRIDE_IN_1X1: False 15 | TRANS_FUNC: "BottleneckWithGN" 16 | STEM_FUNC: "StemWithGN" 17 | FPN: 18 | USE_GN: True # use GN for FPN 19 | RPN: 20 | USE_FPN: True 21 | ANCHOR_STRIDE: (4, 8, 16, 32, 64) 22 | PRE_NMS_TOP_N_TRAIN: 2000 23 | PRE_NMS_TOP_N_TEST: 1000 24 | POST_NMS_TOP_N_TEST: 1000 25 | FPN_POST_NMS_TOP_N_TEST: 1000 26 | ROI_HEADS: 27 | USE_FPN: True 28 | BATCH_SIZE_PER_IMAGE: 512 29 | POSITIVE_FRACTION: 0.25 30 | ROI_BOX_HEAD: 31 | USE_GN: True # use GN for bbox head 32 | POOLER_RESOLUTION: 7 33 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 34 | POOLER_SAMPLING_RATIO: 2 35 | FEATURE_EXTRACTOR: "FPN2MLPFeatureExtractor" 36 | PREDICTOR: "FPNPredictor" 37 | DATASETS: 38 | TRAIN: ("coco_2014_train", "coco_2014_valminusminival") 39 | TEST: ("coco_2014_minival",) 40 | DATALOADER: 41 | SIZE_DIVISIBILITY: 32 42 | SOLVER: 43 | # Assume 8 gpus 44 | BASE_LR: 0.02 45 | WEIGHT_DECAY: 0.0001 46 | STEPS: (210000, 250000) 47 | MAX_ITER: 270000 48 | IMS_PER_BATCH: 16 49 | TEST: 50 | IMS_PER_BATCH: 8 51 | -------------------------------------------------------------------------------- /configs/e2e_mask_rcnn_fbnet_xirb16d_dsmask.yaml: -------------------------------------------------------------------------------- 1 | MODEL: 2 | META_ARCHITECTURE: "GeneralizedRCNN" 3 | BACKBONE: 4 | CONV_BODY: FBNet 5 | FBNET: 6 | ARCH: "xirb16d_dsmask" 7 | BN_TYPE: "bn" 8 | WIDTH_DIVISOR: 8 9 | DW_CONV_SKIP_BN: True 10 | DW_CONV_SKIP_RELU: True 11 | DET_HEAD_LAST_SCALE: -1.0 12 | RPN: 13 | ANCHOR_SIZES: (16, 32, 64, 128, 256) 14 | ANCHOR_STRIDE: (16, ) 15 | BATCH_SIZE_PER_IMAGE: 256 16 | PRE_NMS_TOP_N_TRAIN: 6000 17 | PRE_NMS_TOP_N_TEST: 6000 18 | POST_NMS_TOP_N_TRAIN: 2000 19 | POST_NMS_TOP_N_TEST: 100 20 | RPN_HEAD: FBNet.rpn_head 21 | ROI_HEADS: 22 | BATCH_SIZE_PER_IMAGE: 512 23 | ROI_BOX_HEAD: 24 | POOLER_RESOLUTION: 6 25 | FEATURE_EXTRACTOR: FBNet.roi_head 26 | NUM_CLASSES: 81 27 | ROI_MASK_HEAD: 28 | POOLER_RESOLUTION: 6 29 | FEATURE_EXTRACTOR: FBNet.roi_head_mask 30 | PREDICTOR: "MaskRCNNConv1x1Predictor" 31 | RESOLUTION: 12 32 | SHARE_BOX_FEATURE_EXTRACTOR: False 33 | MASK_ON: True 34 | DATASETS: 35 | TRAIN: ("coco_2014_train", "coco_2014_valminusminival") 36 | TEST: ("coco_2014_minival",) 37 | SOLVER: 38 | BASE_LR: 0.06 39 | WARMUP_FACTOR: 0.1 40 | WEIGHT_DECAY: 0.0001 41 | STEPS: (60000, 80000) 42 | MAX_ITER: 90000 43 | IMS_PER_BATCH: 128 # for 8GPUs 44 | # TEST: 45 | # IMS_PER_BATCH: 8 46 | INPUT: 47 | MIN_SIZE_TRAIN: (320, ) 48 | MAX_SIZE_TRAIN: 640 49 | MIN_SIZE_TEST: 320 50 | MAX_SIZE_TEST: 640 51 | PIXEL_MEAN: [103.53, 116.28, 123.675] 52 | PIXEL_STD: [57.375, 57.12, 58.395] 53 | -------------------------------------------------------------------------------- /configs/e2e_mask_rcnn_fbnet_xirb16d_dsmask_600.yaml: -------------------------------------------------------------------------------- 1 | MODEL: 2 | META_ARCHITECTURE: "GeneralizedRCNN" 3 | BACKBONE: 4 | CONV_BODY: FBNet 5 | FBNET: 6 | ARCH: "xirb16d_dsmask" 7 | BN_TYPE: "bn" 8 | WIDTH_DIVISOR: 8 9 | DW_CONV_SKIP_BN: True 10 | DW_CONV_SKIP_RELU: True 11 | DET_HEAD_LAST_SCALE: 0.0 12 | RPN: 13 | ANCHOR_SIZES: (32, 64, 128, 256, 512) 14 | ANCHOR_STRIDE: (16, ) 15 | BATCH_SIZE_PER_IMAGE: 256 16 | PRE_NMS_TOP_N_TRAIN: 6000 17 | PRE_NMS_TOP_N_TEST: 6000 18 | POST_NMS_TOP_N_TRAIN: 2000 19 | POST_NMS_TOP_N_TEST: 200 20 | RPN_HEAD: FBNet.rpn_head 21 | ROI_HEADS: 22 | BATCH_SIZE_PER_IMAGE: 256 23 | ROI_BOX_HEAD: 24 | POOLER_RESOLUTION: 6 25 | FEATURE_EXTRACTOR: FBNet.roi_head 26 | NUM_CLASSES: 81 27 | ROI_MASK_HEAD: 28 | POOLER_RESOLUTION: 6 29 | FEATURE_EXTRACTOR: FBNet.roi_head_mask 30 | PREDICTOR: "MaskRCNNConv1x1Predictor" 31 | RESOLUTION: 12 32 | SHARE_BOX_FEATURE_EXTRACTOR: False 33 | MASK_ON: True 34 | DATASETS: 35 | TRAIN: ("coco_2014_train", "coco_2014_valminusminival") 36 | TEST: ("coco_2014_minival",) 37 | SOLVER: 38 | BASE_LR: 0.06 39 | WARMUP_FACTOR: 0.1 40 | WEIGHT_DECAY: 0.0001 41 | STEPS: (60000, 80000) 42 | MAX_ITER: 90000 43 | IMS_PER_BATCH: 128 # for 8GPUs 44 | # TEST: 45 | # IMS_PER_BATCH: 8 46 | INPUT: 47 | MIN_SIZE_TRAIN: (600, ) 48 | MAX_SIZE_TRAIN: 1000 49 | MIN_SIZE_TEST: 600 50 | MAX_SIZE_TEST: 1000 51 | PIXEL_MEAN: [103.53, 116.28, 123.675] 52 | PIXEL_STD: [57.375, 57.12, 58.395] 53 | -------------------------------------------------------------------------------- /configs/retinanet/retinanet_X_101_32x8d_FPN_1x.yaml: -------------------------------------------------------------------------------- 1 | MODEL: 2 | META_ARCHITECTURE: "GeneralizedRCNN" 3 | WEIGHT: "catalog://ImageNetPretrained/FAIR/20171220/X-101-32x8d" 4 | RPN_ONLY: True 5 | RETINANET_ON: True 6 | BACKBONE: 7 | CONV_BODY: "R-101-FPN-RETINANET" 8 | RESNETS: 9 | BACKBONE_OUT_CHANNELS: 256 10 | STRIDE_IN_1X1: False 11 | NUM_GROUPS: 32 12 | WIDTH_PER_GROUP: 8 13 | RPN: 14 | USE_FPN: True 15 | FG_IOU_THRESHOLD: 0.5 16 | BG_IOU_THRESHOLD: 0.4 17 | ANCHOR_STRIDE: (4, 8, 16, 32, 64) 18 | PRE_NMS_TOP_N_TRAIN: 2000 19 | PRE_NMS_TOP_N_TEST: 1000 20 | POST_NMS_TOP_N_TEST: 1000 21 | FPN_POST_NMS_TOP_N_TEST: 1000 22 | ROI_HEADS: 23 | USE_FPN: True 24 | BATCH_SIZE_PER_IMAGE: 256 25 | ROI_BOX_HEAD: 26 | POOLER_RESOLUTION: 7 27 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 28 | POOLER_SAMPLING_RATIO: 2 29 | FEATURE_EXTRACTOR: "FPN2MLPFeatureExtractor" 30 | PREDICTOR: "FPNPredictor" 31 | RETINANET: 32 | SCALES_PER_OCTAVE: 3 33 | STRADDLE_THRESH: -1 34 | FG_IOU_THRESHOLD: 0.5 35 | BG_IOU_THRESHOLD: 0.4 36 | DATASETS: 37 | TRAIN: ("coco_2014_train", "coco_2014_valminusminival") 38 | TEST: ("coco_2014_minival",) 39 | INPUT: 40 | MIN_SIZE_TRAIN: (800, ) 41 | MAX_SIZE_TRAIN: 1333 42 | MIN_SIZE_TEST: 800 43 | MAX_SIZE_TEST: 1333 44 | DATALOADER: 45 | SIZE_DIVISIBILITY: 32 46 | SOLVER: 47 | # Assume 4 gpus 48 | BASE_LR: 0.0025 49 | WEIGHT_DECAY: 0.0001 50 | STEPS: (240000, 320000) 51 | MAX_ITER: 360000 52 | IMS_PER_BATCH: 4 53 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/Eigenvalues: -------------------------------------------------------------------------------- 1 | #ifndef EIGEN_EIGENVALUES_MODULE_H 2 | #define EIGEN_EIGENVALUES_MODULE_H 3 | 4 | #include "Core" 5 | 6 | #include "src/Core/util/DisableStupidWarnings.h" 7 | 8 | #include "Cholesky" 9 | #include "Jacobi" 10 | #include "Householder" 11 | #include "LU" 12 | #include "Geometry" 13 | 14 | /** \defgroup Eigenvalues_Module Eigenvalues module 15 | * 16 | * 17 | * 18 | * This module mainly provides various eigenvalue solvers. 19 | * This module also provides some MatrixBase methods, including: 20 | * - MatrixBase::eigenvalues(), 21 | * - MatrixBase::operatorNorm() 22 | * 23 | * \code 24 | * #include 25 | * \endcode 26 | */ 27 | 28 | #include "src/Eigenvalues/Tridiagonalization.h" 29 | #include "src/Eigenvalues/RealSchur.h" 30 | #include "src/Eigenvalues/EigenSolver.h" 31 | #include "src/Eigenvalues/SelfAdjointEigenSolver.h" 32 | #include "src/Eigenvalues/GeneralizedSelfAdjointEigenSolver.h" 33 | #include "src/Eigenvalues/HessenbergDecomposition.h" 34 | #include "src/Eigenvalues/ComplexSchur.h" 35 | #include "src/Eigenvalues/ComplexEigenSolver.h" 36 | #include "src/Eigenvalues/MatrixBaseEigenvalues.h" 37 | #ifdef EIGEN_USE_LAPACKE 38 | #include "src/Eigenvalues/RealSchur_MKL.h" 39 | #include "src/Eigenvalues/ComplexSchur_MKL.h" 40 | #include "src/Eigenvalues/SelfAdjointEigenSolver_MKL.h" 41 | #endif 42 | 43 | #include "src/Core/util/ReenableStupidWarnings.h" 44 | 45 | #endif // EIGEN_EIGENVALUES_MODULE_H 46 | /* vim: set filetype=cpp et sw=2 ts=2 ai: */ 47 | -------------------------------------------------------------------------------- /configs/gn_baselines/e2e_faster_rcnn_R_50_FPN_Xconv1fc_1x_gn.yaml: -------------------------------------------------------------------------------- 1 | INPUT: 2 | MIN_SIZE_TRAIN: (800,) 3 | MAX_SIZE_TRAIN: 1333 4 | MIN_SIZE_TEST: 800 5 | MAX_SIZE_TEST: 1333 6 | MODEL: 7 | META_ARCHITECTURE: "GeneralizedRCNN" 8 | WEIGHT: "catalog://ImageNetPretrained/MSRA/R-50-GN" 9 | BACKBONE: 10 | CONV_BODY: "R-50-FPN" 11 | RESNETS: # use GN for backbone 12 | BACKBONE_OUT_CHANNELS: 256 13 | STRIDE_IN_1X1: False 14 | TRANS_FUNC: "BottleneckWithGN" 15 | STEM_FUNC: "StemWithGN" 16 | FPN: 17 | USE_GN: True # use GN for FPN 18 | RPN: 19 | USE_FPN: True 20 | ANCHOR_STRIDE: (4, 8, 16, 32, 64) 21 | PRE_NMS_TOP_N_TRAIN: 2000 22 | PRE_NMS_TOP_N_TEST: 1000 23 | POST_NMS_TOP_N_TEST: 1000 24 | FPN_POST_NMS_TOP_N_TEST: 1000 25 | ROI_HEADS: 26 | USE_FPN: True 27 | BATCH_SIZE_PER_IMAGE: 512 28 | POSITIVE_FRACTION: 0.25 29 | ROI_BOX_HEAD: 30 | USE_GN: True # use GN for bbox head 31 | POOLER_RESOLUTION: 7 32 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 33 | POOLER_SAMPLING_RATIO: 2 34 | CONV_HEAD_DIM: 256 35 | NUM_STACKED_CONVS: 4 36 | FEATURE_EXTRACTOR: "FPNXconv1fcFeatureExtractor" 37 | PREDICTOR: "FPNPredictor" 38 | DATASETS: 39 | TRAIN: ("coco_2014_train", "coco_2014_valminusminival") 40 | TEST: ("coco_2014_minival",) 41 | DATALOADER: 42 | SIZE_DIVISIBILITY: 32 43 | SOLVER: 44 | # Assume 8 gpus 45 | BASE_LR: 0.02 46 | WEIGHT_DECAY: 0.0001 47 | STEPS: (60000, 80000) 48 | MAX_ITER: 90000 49 | IMS_PER_BATCH: 16 50 | TEST: 51 | IMS_PER_BATCH: 8 52 | -------------------------------------------------------------------------------- /configs/BBTP/e2e_mask_rcnn_R_101_FPN_4x_voc_aug_cocostyle.yaml: -------------------------------------------------------------------------------- 1 | MODEL: 2 | META_ARCHITECTURE: "GeneralizedRCNN" 3 | WEIGHT: "catalog://ImageNetPretrained/MSRA/R-101" 4 | BACKBONE: 5 | CONV_BODY: "R-101-FPN" 6 | RESNETS: 7 | BACKBONE_OUT_CHANNELS: 256 8 | RPN: 9 | USE_FPN: True 10 | ANCHOR_STRIDE: (4, 8, 16, 32, 64) 11 | PRE_NMS_TOP_N_TRAIN: 2000 12 | PRE_NMS_TOP_N_TEST: 1000 13 | POST_NMS_TOP_N_TEST: 1000 14 | FPN_POST_NMS_TOP_N_TEST: 1000 15 | ROI_HEADS: 16 | USE_FPN: True 17 | SCORE_THRESH: 0.05 18 | FG_IOU_THRESHOLD: 0.5 19 | BG_IOU_THRESHOLD: 0.5 20 | ROI_BOX_HEAD: 21 | POOLER_RESOLUTION: 7 22 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 23 | POOLER_SAMPLING_RATIO: 2 24 | FEATURE_EXTRACTOR: "FPN2MLPFeatureExtractor" 25 | PREDICTOR: "FPNPredictor" 26 | NUM_CLASSES: 21 27 | ROI_MASK_HEAD: 28 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 29 | FEATURE_EXTRACTOR: "MaskRCNNFPNFeatureExtractor" 30 | PREDICTOR: "MaskRCNNC4Predictor" 31 | POOLER_RESOLUTION: 14 32 | POOLER_SAMPLING_RATIO: 2 33 | RESOLUTION: 28 34 | SHARE_BOX_FEATURE_EXTRACTOR: False 35 | POSTPROCESS_MASKS: False 36 | POSTPROCESS_MASKS_THRESHOLD: 0.5 37 | MASK_ON: True 38 | DATASETS: 39 | TRAIN: ("voc_2012_aug_train_cocostyle", ) 40 | TEST: ("voc_2012_val_cocostyle",) 41 | DATALOADER: 42 | SIZE_DIVISIBILITY: 32 43 | SOLVER: 44 | BASE_LR: 0.01 45 | WEIGHT_DECAY: 0.0001 46 | STEPS: (15000, ) 47 | MAX_ITER: 20000 48 | IMS_PER_BATCH: 8 49 | TEST: 50 | IMS_PER_BATCH: 8 51 | SAVE_PRED_AS_MAT: True -------------------------------------------------------------------------------- /configs/quick_schedules/e2e_keypoint_rcnn_R_50_FPN_quick.yaml: -------------------------------------------------------------------------------- 1 | MODEL: 2 | META_ARCHITECTURE: "GeneralizedRCNN" 3 | WEIGHT: "catalog://ImageNetPretrained/MSRA/R-50" 4 | BACKBONE: 5 | CONV_BODY: "R-50-FPN" 6 | RESNETS: 7 | BACKBONE_OUT_CHANNELS: 256 8 | RPN: 9 | USE_FPN: True 10 | ANCHOR_STRIDE: (4, 8, 16, 32, 64) 11 | PRE_NMS_TOP_N_TRAIN: 2000 12 | PRE_NMS_TOP_N_TEST: 1000 13 | POST_NMS_TOP_N_TEST: 1000 14 | FPN_POST_NMS_TOP_N_TEST: 1000 15 | ROI_HEADS: 16 | USE_FPN: True 17 | BATCH_SIZE_PER_IMAGE: 256 18 | ROI_BOX_HEAD: 19 | POOLER_RESOLUTION: 7 20 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 21 | POOLER_SAMPLING_RATIO: 2 22 | FEATURE_EXTRACTOR: "FPN2MLPFeatureExtractor" 23 | PREDICTOR: "FPNPredictor" 24 | NUM_CLASSES: 2 25 | ROI_KEYPOINT_HEAD: 26 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 27 | FEATURE_EXTRACTOR: "KeypointRCNNFeatureExtractor" 28 | PREDICTOR: "KeypointRCNNPredictor" 29 | POOLER_RESOLUTION: 14 30 | POOLER_SAMPLING_RATIO: 2 31 | RESOLUTION: 56 32 | SHARE_BOX_FEATURE_EXTRACTOR: False 33 | KEYPOINT_ON: True 34 | DATASETS: 35 | TRAIN: ("keypoints_coco_2014_minival",) 36 | TEST: ("keypoints_coco_2014_minival",) 37 | INPUT: 38 | MIN_SIZE_TRAIN: (640, 672, 704, 736, 768, 800) 39 | MAX_SIZE_TRAIN: 1000 40 | MIN_SIZE_TEST: 800 41 | MAX_SIZE_TEST: 1000 42 | DATALOADER: 43 | SIZE_DIVISIBILITY: 32 44 | SOLVER: 45 | BASE_LR: 0.005 46 | WEIGHT_DECAY: 0.0001 47 | STEPS: (1500,) 48 | MAX_ITER: 2000 49 | IMS_PER_BATCH: 4 50 | TEST: 51 | IMS_PER_BATCH: 2 52 | -------------------------------------------------------------------------------- /maskrcnn_benchmark/utils/registry.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. 2 | 3 | 4 | def _register_generic(module_dict, module_name, module): 5 | assert module_name not in module_dict 6 | module_dict[module_name] = module 7 | 8 | 9 | class Registry(dict): 10 | ''' 11 | A helper class for managing registering modules, it extends a dictionary 12 | and provides a register functions. 13 | 14 | Eg. creeting a registry: 15 | some_registry = Registry({"default": default_module}) 16 | 17 | There're two ways of registering new modules: 18 | 1): normal way is just calling register function: 19 | def foo(): 20 | ... 21 | some_registry.register("foo_module", foo) 22 | 2): used as decorator when declaring the module: 23 | @some_registry.register("foo_module") 24 | @some_registry.register("foo_modeul_nickname") 25 | def foo(): 26 | ... 27 | 28 | Access of module is just like using a dictionary, eg: 29 | f = some_registry["foo_modeul"] 30 | ''' 31 | def __init__(self, *args, **kwargs): 32 | super(Registry, self).__init__(*args, **kwargs) 33 | 34 | def register(self, module_name, module=None): 35 | # used as function call 36 | if module is not None: 37 | _register_generic(self, module_name, module) 38 | return 39 | 40 | # used as decorator 41 | def register_fn(fn): 42 | _register_generic(self, module_name, fn) 43 | return fn 44 | 45 | return register_fn 46 | -------------------------------------------------------------------------------- /configs/gn_baselines/scratch_e2e_faster_rcnn_R_50_FPN_Xconv1fc_3x_gn.yaml: -------------------------------------------------------------------------------- 1 | INPUT: 2 | MIN_SIZE_TRAIN: (800,) 3 | MAX_SIZE_TRAIN: 1333 4 | MIN_SIZE_TEST: 800 5 | MAX_SIZE_TEST: 1333 6 | MODEL: 7 | META_ARCHITECTURE: "GeneralizedRCNN" 8 | WEIGHT: "" # no pretrained model 9 | BACKBONE: 10 | CONV_BODY: "R-50-FPN" 11 | FREEZE_CONV_BODY_AT: 0 # finetune all layers 12 | RESNETS: # use GN for backbone 13 | BACKBONE_OUT_CHANNELS: 256 14 | STRIDE_IN_1X1: False 15 | TRANS_FUNC: "BottleneckWithGN" 16 | STEM_FUNC: "StemWithGN" 17 | FPN: 18 | USE_GN: True # use GN for FPN 19 | RPN: 20 | USE_FPN: True 21 | ANCHOR_STRIDE: (4, 8, 16, 32, 64) 22 | PRE_NMS_TOP_N_TRAIN: 2000 23 | PRE_NMS_TOP_N_TEST: 1000 24 | POST_NMS_TOP_N_TEST: 1000 25 | FPN_POST_NMS_TOP_N_TEST: 1000 26 | ROI_HEADS: 27 | USE_FPN: True 28 | BATCH_SIZE_PER_IMAGE: 512 29 | POSITIVE_FRACTION: 0.25 30 | ROI_BOX_HEAD: 31 | USE_GN: True # use GN for bbox head 32 | POOLER_RESOLUTION: 7 33 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 34 | POOLER_SAMPLING_RATIO: 2 35 | CONV_HEAD_DIM: 256 36 | NUM_STACKED_CONVS: 4 37 | FEATURE_EXTRACTOR: "FPNXconv1fcFeatureExtractor" 38 | PREDICTOR: "FPNPredictor" 39 | DATASETS: 40 | TRAIN: ("coco_2014_train", "coco_2014_valminusminival") 41 | TEST: ("coco_2014_minival",) 42 | DATALOADER: 43 | SIZE_DIVISIBILITY: 32 44 | SOLVER: 45 | # Assume 8 gpus 46 | BASE_LR: 0.02 47 | WEIGHT_DECAY: 0.0001 48 | STEPS: (210000, 250000) 49 | MAX_ITER: 270000 50 | IMS_PER_BATCH: 16 51 | TEST: 52 | IMS_PER_BATCH: 8 53 | -------------------------------------------------------------------------------- /configs/BBTP/e2e_mask_rcnn_R_101_FPN_4x_voc_coco_aug_cocostyle.yaml: -------------------------------------------------------------------------------- 1 | MODEL: 2 | META_ARCHITECTURE: "GeneralizedRCNN" 3 | WEIGHT: "catalog://ImageNetPretrained/MSRA/R-101" 4 | BACKBONE: 5 | CONV_BODY: "R-101-FPN" 6 | RESNETS: 7 | BACKBONE_OUT_CHANNELS: 256 8 | RPN: 9 | USE_FPN: True 10 | ANCHOR_STRIDE: (4, 8, 16, 32, 64) 11 | PRE_NMS_TOP_N_TRAIN: 2000 12 | PRE_NMS_TOP_N_TEST: 1000 13 | POST_NMS_TOP_N_TEST: 1000 14 | FPN_POST_NMS_TOP_N_TEST: 1000 15 | ROI_HEADS: 16 | USE_FPN: True 17 | SCORE_THRESH: 0.05 18 | FG_IOU_THRESHOLD: 0.5 19 | BG_IOU_THRESHOLD: 0.5 20 | ROI_BOX_HEAD: 21 | POOLER_RESOLUTION: 7 22 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 23 | POOLER_SAMPLING_RATIO: 2 24 | FEATURE_EXTRACTOR: "FPN2MLPFeatureExtractor" 25 | PREDICTOR: "FPNPredictor" 26 | NUM_CLASSES: 21 # 90 # 21 27 | ROI_MASK_HEAD: 28 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 29 | FEATURE_EXTRACTOR: "MaskRCNNFPNFeatureExtractor" 30 | PREDICTOR: "MaskRCNNC4Predictor" 31 | POOLER_RESOLUTION: 14 32 | POOLER_SAMPLING_RATIO: 2 33 | RESOLUTION: 28 34 | SHARE_BOX_FEATURE_EXTRACTOR: False 35 | POSTPROCESS_MASKS: False 36 | POSTPROCESS_MASKS_THRESHOLD: 0.5 37 | MASK_ON: True 38 | DATASETS: 39 | TRAIN: ("voc_2012_coco_aug_train_cocostyle", ) 40 | TEST: ("voc_2012_val_cocostyle",) 41 | DATALOADER: 42 | SIZE_DIVISIBILITY: 32 43 | SOLVER: 44 | BASE_LR: 0.01 45 | WEIGHT_DECAY: 0.0001 46 | STEPS: (30000, ) 47 | MAX_ITER: 52500 48 | IMS_PER_BATCH: 8 49 | TEST: 50 | IMS_PER_BATCH: 8 51 | SAVE_PRED_AS_MAT: False -------------------------------------------------------------------------------- /configs/quick_schedules/e2e_mask_rcnn_X_101_32x8d_FPN_quick.yaml: -------------------------------------------------------------------------------- 1 | MODEL: 2 | META_ARCHITECTURE: "GeneralizedRCNN" 3 | WEIGHT: "catalog://ImageNetPretrained/FAIR/20171220/X-101-32x8d" 4 | BACKBONE: 5 | CONV_BODY: "R-101-FPN" 6 | RESNETS: 7 | BACKBONE_OUT_CHANNELS: 256 8 | STRIDE_IN_1X1: False 9 | NUM_GROUPS: 32 10 | WIDTH_PER_GROUP: 8 11 | RPN: 12 | USE_FPN: True 13 | ANCHOR_STRIDE: (4, 8, 16, 32, 64) 14 | PRE_NMS_TOP_N_TRAIN: 2000 15 | PRE_NMS_TOP_N_TEST: 1000 16 | POST_NMS_TOP_N_TEST: 1000 17 | FPN_POST_NMS_TOP_N_TEST: 1000 18 | ROI_HEADS: 19 | USE_FPN: True 20 | BATCH_SIZE_PER_IMAGE: 256 21 | ROI_BOX_HEAD: 22 | POOLER_RESOLUTION: 7 23 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 24 | POOLER_SAMPLING_RATIO: 2 25 | FEATURE_EXTRACTOR: "FPN2MLPFeatureExtractor" 26 | PREDICTOR: "FPNPredictor" 27 | ROI_MASK_HEAD: 28 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 29 | FEATURE_EXTRACTOR: "MaskRCNNFPNFeatureExtractor" 30 | PREDICTOR: "MaskRCNNC4Predictor" 31 | POOLER_RESOLUTION: 14 32 | POOLER_SAMPLING_RATIO: 2 33 | RESOLUTION: 28 34 | SHARE_BOX_FEATURE_EXTRACTOR: False 35 | MASK_ON: True 36 | DATASETS: 37 | TRAIN: ("coco_2014_minival",) 38 | TEST: ("coco_2014_minival",) 39 | INPUT: 40 | MIN_SIZE_TRAIN: (600,) 41 | MAX_SIZE_TRAIN: 1000 42 | MIN_SIZE_TEST: 800 43 | MAX_SIZE_TEST: 1000 44 | DATALOADER: 45 | SIZE_DIVISIBILITY: 32 46 | SOLVER: 47 | BASE_LR: 0.005 48 | WEIGHT_DECAY: 0.0001 49 | STEPS: (1500,) 50 | MAX_ITER: 2000 51 | IMS_PER_BATCH: 2 52 | TEST: 53 | IMS_PER_BATCH: 2 54 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/PaStiXSupport: -------------------------------------------------------------------------------- 1 | #ifndef EIGEN_PASTIXSUPPORT_MODULE_H 2 | #define EIGEN_PASTIXSUPPORT_MODULE_H 3 | 4 | #include "SparseCore" 5 | 6 | #include "src/Core/util/DisableStupidWarnings.h" 7 | 8 | #include 9 | extern "C" { 10 | #include 11 | #include 12 | } 13 | 14 | #ifdef complex 15 | #undef complex 16 | #endif 17 | 18 | /** \ingroup Support_modules 19 | * \defgroup PaStiXSupport_Module PaStiXSupport module 20 | * 21 | * This module provides an interface to the PaSTiX library. 22 | * PaSTiX is a general \b supernodal, \b parallel and \b opensource sparse solver. 23 | * It provides the two following main factorization classes: 24 | * - class PastixLLT : a supernodal, parallel LLt Cholesky factorization. 25 | * - class PastixLDLT: a supernodal, parallel LDLt Cholesky factorization. 26 | * - class PastixLU : a supernodal, parallel LU factorization (optimized for a symmetric pattern). 27 | * 28 | * \code 29 | * #include 30 | * \endcode 31 | * 32 | * In order to use this module, the PaSTiX headers must be accessible from the include paths, and your binary must be linked to the PaSTiX library and its dependencies. 33 | * The dependencies depend on how PaSTiX has been compiled. 34 | * For a cmake based project, you can use our FindPaSTiX.cmake module to help you in this task. 35 | * 36 | */ 37 | 38 | #include "src/misc/Solve.h" 39 | #include "src/misc/SparseSolve.h" 40 | 41 | #include "src/PaStiXSupport/PaStiXSupport.h" 42 | 43 | 44 | #include "src/Core/util/ReenableStupidWarnings.h" 45 | 46 | #endif // EIGEN_PASTIXSUPPORT_MODULE_H 47 | -------------------------------------------------------------------------------- /demo/README.md: -------------------------------------------------------------------------------- 1 | ## Webcam and Jupyter notebook demo 2 | 3 | This folder contains a simple webcam demo that illustrates how you can use `maskrcnn_benchmark` for inference. 4 | 5 | 6 | ### With your preferred environment 7 | 8 | You can start it by running it from this folder, using one of the following commands: 9 | ```bash 10 | # by default, it runs on the GPU 11 | # for best results, use min-image-size 800 12 | python webcam.py --min-image-size 800 13 | # can also run it on the CPU 14 | python webcam.py --min-image-size 300 MODEL.DEVICE cpu 15 | # or change the model that you want to use 16 | python webcam.py --config-file ../configs/caffe2/e2e_mask_rcnn_R_101_FPN_1x_caffe2.yaml --min-image-size 300 MODEL.DEVICE cpu 17 | # in order to see the probability heatmaps, pass --show-mask-heatmaps 18 | python webcam.py --min-image-size 300 --show-mask-heatmaps MODEL.DEVICE cpu 19 | ``` 20 | 21 | ### With Docker 22 | 23 | Build the image with the tag `maskrcnn-benchmark` (check [INSTALL.md](../INSTALL.md) for instructions) 24 | 25 | Adjust permissions of the X server host (be careful with this step, refer to 26 | [here](http://wiki.ros.org/docker/Tutorials/GUI) for alternatives) 27 | 28 | ```bash 29 | xhost + 30 | ``` 31 | 32 | Then run a container with the demo: 33 | 34 | ``` 35 | docker run --rm -it \ 36 | -e DISPLAY=${DISPLAY} \ 37 | --privileged \ 38 | -v /tmp/.X11-unix:/tmp/.X11-unix \ 39 | --device=/dev/video0:/dev/video0 \ 40 | --ipc=host maskrcnn-benchmark \ 41 | python demo/webcam.py --min-image-size 300 \ 42 | --config-file configs/caffe2/e2e_mask_rcnn_R_50_FPN_1x_caffe2.yaml 43 | ``` 44 | 45 | **DISCLAIMER:** *This was tested for an Ubuntu 16.04 machine, 46 | the volume mapping may vary depending on your platform* 47 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/src/Eigen2Support/TriangularSolver.h: -------------------------------------------------------------------------------- 1 | // This file is part of Eigen, a lightweight C++ template library 2 | // for linear algebra. 3 | // 4 | // Copyright (C) 2010 Gael Guennebaud 5 | // 6 | // This Source Code Form is subject to the terms of the Mozilla 7 | // Public License v. 2.0. If a copy of the MPL was not distributed 8 | // with this file, You can obtain one at http://mozilla.org/MPL/2.0/. 9 | 10 | #ifndef EIGEN_TRIANGULAR_SOLVER2_H 11 | #define EIGEN_TRIANGULAR_SOLVER2_H 12 | 13 | namespace Eigen { 14 | 15 | const unsigned int UnitDiagBit = UnitDiag; 16 | const unsigned int SelfAdjointBit = SelfAdjoint; 17 | const unsigned int UpperTriangularBit = Upper; 18 | const unsigned int LowerTriangularBit = Lower; 19 | 20 | const unsigned int UpperTriangular = Upper; 21 | const unsigned int LowerTriangular = Lower; 22 | const unsigned int UnitUpperTriangular = UnitUpper; 23 | const unsigned int UnitLowerTriangular = UnitLower; 24 | 25 | template 26 | template 27 | typename ExpressionType::PlainObject 28 | Flagged::solveTriangular(const MatrixBase& other) const 29 | { 30 | return m_matrix.template triangularView().solve(other.derived()); 31 | } 32 | 33 | template 34 | template 35 | void Flagged::solveTriangularInPlace(const MatrixBase& other) const 36 | { 37 | m_matrix.template triangularView().solveInPlace(other.derived()); 38 | } 39 | 40 | } // end namespace Eigen 41 | 42 | #endif // EIGEN_TRIANGULAR_SOLVER2_H 43 | -------------------------------------------------------------------------------- /CONTRIBUTING.md: -------------------------------------------------------------------------------- 1 | # Contributing to Mask-RCNN Benchmark 2 | We want to make contributing to this project as easy and transparent as 3 | possible. 4 | 5 | ## Our Development Process 6 | Minor changes and improvements will be released on an ongoing basis. Larger changes (e.g., changesets implementing a new paper) will be released on a more periodic basis. 7 | 8 | ## Pull Requests 9 | We actively welcome your pull requests. 10 | 11 | 1. Fork the repo and create your branch from `master`. 12 | 2. If you've added code that should be tested, add tests. 13 | 3. If you've changed APIs, update the documentation. 14 | 4. Ensure the test suite passes. 15 | 5. Make sure your code lints. 16 | 6. If you haven't already, complete the Contributor License Agreement ("CLA"). 17 | 18 | ## Contributor License Agreement ("CLA") 19 | In order to accept your pull request, we need you to submit a CLA. You only need 20 | to do this once to work on any of Facebook's open source projects. 21 | 22 | Complete your CLA here: 23 | 24 | ## Issues 25 | We use GitHub issues to track public bugs. Please ensure your description is 26 | clear and has sufficient instructions to be able to reproduce the issue. 27 | 28 | Facebook has a [bounty program](https://www.facebook.com/whitehat/) for the safe 29 | disclosure of security bugs. In those cases, please go through the process 30 | outlined on that page and do not file a public issue. 31 | 32 | ## Coding Style 33 | * 4 spaces for indentation rather than tabs 34 | * 80 character line length 35 | * PEP8 formatting following [Black](https://black.readthedocs.io/en/stable/) 36 | 37 | ## License 38 | By contributing to Mask-RCNN Benchmark, you agree that your contributions will be licensed 39 | under the LICENSE file in the root directory of this source tree. 40 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/src/SparseCore/SparseRedux.h: -------------------------------------------------------------------------------- 1 | // This file is part of Eigen, a lightweight C++ template library 2 | // for linear algebra. 3 | // 4 | // Copyright (C) 2008 Gael Guennebaud 5 | // 6 | // This Source Code Form is subject to the terms of the Mozilla 7 | // Public License v. 2.0. If a copy of the MPL was not distributed 8 | // with this file, You can obtain one at http://mozilla.org/MPL/2.0/. 9 | 10 | #ifndef EIGEN_SPARSEREDUX_H 11 | #define EIGEN_SPARSEREDUX_H 12 | 13 | namespace Eigen { 14 | 15 | template 16 | typename internal::traits::Scalar 17 | SparseMatrixBase::sum() const 18 | { 19 | eigen_assert(rows()>0 && cols()>0 && "you are using a non initialized matrix"); 20 | Scalar res(0); 21 | for (Index j=0; j 28 | typename internal::traits >::Scalar 29 | SparseMatrix<_Scalar,_Options,_Index>::sum() const 30 | { 31 | eigen_assert(rows()>0 && cols()>0 && "you are using a non initialized matrix"); 32 | return Matrix::Map(&m_data.value(0), m_data.size()).sum(); 33 | } 34 | 35 | template 36 | typename internal::traits >::Scalar 37 | SparseVector<_Scalar,_Options,_Index>::sum() const 38 | { 39 | eigen_assert(rows()>0 && cols()>0 && "you are using a non initialized matrix"); 40 | return Matrix::Map(&m_data.value(0), m_data.size()).sum(); 41 | } 42 | 43 | } // end namespace Eigen 44 | 45 | #endif // EIGEN_SPARSEREDUX_H 46 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/IterativeLinearSolvers: -------------------------------------------------------------------------------- 1 | #ifndef EIGEN_ITERATIVELINEARSOLVERS_MODULE_H 2 | #define EIGEN_ITERATIVELINEARSOLVERS_MODULE_H 3 | 4 | #include "SparseCore" 5 | #include "OrderingMethods" 6 | 7 | #include "src/Core/util/DisableStupidWarnings.h" 8 | 9 | /** \ingroup Sparse_modules 10 | * \defgroup IterativeLinearSolvers_Module IterativeLinearSolvers module 11 | * 12 | * This module currently provides iterative methods to solve problems of the form \c A \c x = \c b, where \c A is a squared matrix, usually very large and sparse. 13 | * Those solvers are accessible via the following classes: 14 | * - ConjugateGradient for selfadjoint (hermitian) matrices, 15 | * - BiCGSTAB for general square matrices. 16 | * 17 | * These iterative solvers are associated with some preconditioners: 18 | * - IdentityPreconditioner - not really useful 19 | * - DiagonalPreconditioner - also called JAcobi preconditioner, work very well on diagonal dominant matrices. 20 | * - IncompleteILUT - incomplete LU factorization with dual thresholding 21 | * 22 | * Such problems can also be solved using the direct sparse decomposition modules: SparseCholesky, CholmodSupport, UmfPackSupport, SuperLUSupport. 23 | * 24 | * \code 25 | * #include 26 | * \endcode 27 | */ 28 | 29 | #include "src/misc/Solve.h" 30 | #include "src/misc/SparseSolve.h" 31 | 32 | #include "src/IterativeLinearSolvers/IterativeSolverBase.h" 33 | #include "src/IterativeLinearSolvers/BasicPreconditioners.h" 34 | #include "src/IterativeLinearSolvers/ConjugateGradient.h" 35 | #include "src/IterativeLinearSolvers/BiCGSTAB.h" 36 | #include "src/IterativeLinearSolvers/IncompleteLUT.h" 37 | 38 | #include "src/Core/util/ReenableStupidWarnings.h" 39 | 40 | #endif // EIGEN_ITERATIVELINEARSOLVERS_MODULE_H 41 | -------------------------------------------------------------------------------- /maskrcnn_benchmark/csrc/ROIPool.h: -------------------------------------------------------------------------------- 1 | // Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. 2 | #pragma once 3 | 4 | #include "cpu/vision.h" 5 | 6 | #ifdef WITH_CUDA 7 | #include "cuda/vision.h" 8 | #endif 9 | 10 | 11 | std::tuple ROIPool_forward(const at::Tensor& input, 12 | const at::Tensor& rois, 13 | const float spatial_scale, 14 | const int pooled_height, 15 | const int pooled_width) { 16 | if (input.type().is_cuda()) { 17 | #ifdef WITH_CUDA 18 | return ROIPool_forward_cuda(input, rois, spatial_scale, pooled_height, pooled_width); 19 | #else 20 | AT_ERROR("Not compiled with GPU support"); 21 | #endif 22 | } 23 | AT_ERROR("Not implemented on the CPU"); 24 | } 25 | 26 | at::Tensor ROIPool_backward(const at::Tensor& grad, 27 | const at::Tensor& input, 28 | const at::Tensor& rois, 29 | const at::Tensor& argmax, 30 | const float spatial_scale, 31 | const int pooled_height, 32 | const int pooled_width, 33 | const int batch_size, 34 | const int channels, 35 | const int height, 36 | const int width) { 37 | if (grad.type().is_cuda()) { 38 | #ifdef WITH_CUDA 39 | return ROIPool_backward_cuda(grad, input, rois, argmax, spatial_scale, pooled_height, pooled_width, batch_size, channels, height, width); 40 | #else 41 | AT_ERROR("Not compiled with GPU support"); 42 | #endif 43 | } 44 | AT_ERROR("Not implemented on the CPU"); 45 | } 46 | 47 | 48 | 49 | -------------------------------------------------------------------------------- /maskrcnn_benchmark/csrc/ROIAlign.h: -------------------------------------------------------------------------------- 1 | // Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. 2 | #pragma once 3 | 4 | #include "cpu/vision.h" 5 | 6 | #ifdef WITH_CUDA 7 | #include "cuda/vision.h" 8 | #endif 9 | 10 | // Interface for Python 11 | at::Tensor ROIAlign_forward(const at::Tensor& input, 12 | const at::Tensor& rois, 13 | const float spatial_scale, 14 | const int pooled_height, 15 | const int pooled_width, 16 | const int sampling_ratio) { 17 | if (input.type().is_cuda()) { 18 | #ifdef WITH_CUDA 19 | return ROIAlign_forward_cuda(input, rois, spatial_scale, pooled_height, pooled_width, sampling_ratio); 20 | #else 21 | AT_ERROR("Not compiled with GPU support"); 22 | #endif 23 | } 24 | return ROIAlign_forward_cpu(input, rois, spatial_scale, pooled_height, pooled_width, sampling_ratio); 25 | } 26 | 27 | at::Tensor ROIAlign_backward(const at::Tensor& grad, 28 | const at::Tensor& rois, 29 | const float spatial_scale, 30 | const int pooled_height, 31 | const int pooled_width, 32 | const int batch_size, 33 | const int channels, 34 | const int height, 35 | const int width, 36 | const int sampling_ratio) { 37 | if (grad.type().is_cuda()) { 38 | #ifdef WITH_CUDA 39 | return ROIAlign_backward_cuda(grad, rois, spatial_scale, pooled_height, pooled_width, batch_size, channels, height, width, sampling_ratio); 40 | #else 41 | AT_ERROR("Not compiled with GPU support"); 42 | #endif 43 | } 44 | AT_ERROR("Not implemented on the CPU"); 45 | } 46 | 47 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/maxflow-v3.03.src/CHANGES.TXT: -------------------------------------------------------------------------------- 1 | List of changes from version 3.02: 2 | 3 | - put under GPL license 4 | 5 | List of changes from version 3.01: 6 | 7 | - fixed a bug: using add_node() or add_edge() after the first maxflow() with the reuse_trees option 8 | could have caused segmentation fault (if nodes or arcs are reallocated). Thanks to Jan Lellmann for pointing out this bug. 9 | - updated block.h to suppress compilation warnings 10 | 11 | List of changes from version 3.0: 12 | - Moved line 13 | #include "instances.inc" 14 | to the end of cpp files to make it compile under GNU c++ compilers 4.2(?) and above 15 | 16 | List of changes from version 2.2: 17 | 18 | - Added functions for accessing graph structure, residual capacities, etc. 19 | (They are needed for implementing maxflow-based algorithms such as primal-dual algorithm for convex MRFs.) 20 | - Added option of reusing trees. 21 | - node_id's are now integers starting from 0. Thus, it is not necessary to store node_id's in a separate array. 22 | - Capacity types are now templated. 23 | - Fixed bug in block.h. (After Block::Reset, ScanFirst() and ScanNext() did not work properly). 24 | - Implementation with a forward star representation of the graph is no longer supported. (It needs less memory, but slightly slower than adjacency list representation.) If you still wish to use it, download version 2.2. 25 | - Note: version 3.0 is released under a different license than version 2.2. 26 | 27 | List of changes from version 2.1: 28 | 29 | - Put the code under GPL license 30 | 31 | List of changes from version 2.02: 32 | 33 | - Fixed a bug in the implementation that uses forward star representation 34 | 35 | List of changes from version 2.01: 36 | 37 | - Added new interface function - Graph::add_tweights(Node_id, captype, captype) 38 | (necessary for the "ENERGY" software package) 39 | 40 | -------------------------------------------------------------------------------- /matlab/DenseCRF/README.md: -------------------------------------------------------------------------------- 1 | A MATLAB wrapper for solving DenseCRF problems [1,2]. 2 | The code uses the c++ library provided with [2]. 3 | 4 | Getting started 5 | -- 6 | * Setup a C++ compiler in MATLAB using `mex -setup`, you will need to have a supported compiler installed, see [https://www.mathworks.com/support/compilers.html](https://www.mathworks.com/support/compilers.html) for a list 7 | * To solve a general problem see examples/example.m 8 | * To perform segmentation on the MSRC-21 database using the unary potentials from [http://graphics.stanford.edu/projects/densecrf/unary/](http://graphics.stanford.edu/projects/densecrf/unary/) see examples/example_MSRC.m 9 | 10 | ![Image and result](screenshot/screenshot.png) 11 | 12 | Included solvers 13 | -- 14 | * Mean field approximation, using approximate filtering [2] 15 | * Mean field approximation, performing all summations explicitly (slow) 16 | * TRW-S [3] 17 | * Graph cuts [4] (only works for 2 label problems) 18 | 19 | 20 | References 21 | -- 22 | 1. __Efficient Inference in Fully Connected CRFs with Gaussian Edge Potentials__.
23 | Conference on Neural Information Processing Systems (NIPS), 2011.
24 | _Philipp Krähenbühl and Vladlen Koltun_. 25 | 26 | 2. __Parameter Learning and Convergent Inference for Dense Random Fields__.
27 | International Conference on Machine Learning (ICML), 2013.
28 | _Philipp Krähenbühl and Vladlen Koltun_. 29 | 30 | 3. __Convergent Tree-reweighted Message Passing for Energy Minimization__.
31 | IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI), 2006.
32 | _Vladimir Kolmogorov_. 33 | 34 | 4. __An Experimental Comparison of Min-Cut/Max-Flow Algorithms for Energy Minimization in Computer Vision__.
35 | IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI), 2004
36 | _Yuri Boykov and Vladimir Kolmogorov_. 37 | -------------------------------------------------------------------------------- /maskrcnn_benchmark/modeling/rpn/utils.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. 2 | """ 3 | Utility functions minipulating the prediction layers 4 | """ 5 | 6 | from ..utils import cat 7 | 8 | import torch 9 | 10 | def permute_and_flatten(layer, N, A, C, H, W): 11 | layer = layer.view(N, -1, C, H, W) 12 | layer = layer.permute(0, 3, 4, 1, 2) 13 | layer = layer.reshape(N, -1, C) 14 | return layer 15 | 16 | 17 | def concat_box_prediction_layers(box_cls, box_regression): 18 | box_cls_flattened = [] 19 | box_regression_flattened = [] 20 | # for each feature level, permute the outputs to make them be in the 21 | # same format as the labels. Note that the labels are computed for 22 | # all feature levels concatenated, so we keep the same representation 23 | # for the objectness and the box_regression 24 | for box_cls_per_level, box_regression_per_level in zip( 25 | box_cls, box_regression 26 | ): 27 | N, AxC, H, W = box_cls_per_level.shape 28 | Ax4 = box_regression_per_level.shape[1] 29 | A = Ax4 // 4 30 | C = AxC // A 31 | box_cls_per_level = permute_and_flatten( 32 | box_cls_per_level, N, A, C, H, W 33 | ) 34 | box_cls_flattened.append(box_cls_per_level) 35 | 36 | box_regression_per_level = permute_and_flatten( 37 | box_regression_per_level, N, A, 4, H, W 38 | ) 39 | box_regression_flattened.append(box_regression_per_level) 40 | # concatenate on the first dimension (representing the feature levels), to 41 | # take into account the way the labels were generated (with all feature maps 42 | # being concatenated as well) 43 | box_cls = cat(box_cls_flattened, dim=1).reshape(-1, C) 44 | box_regression = cat(box_regression_flattened, dim=1).reshape(-1, 4) 45 | return box_cls, box_regression 46 | -------------------------------------------------------------------------------- /configs/gn_baselines/e2e_mask_rcnn_R_50_FPN_1x_gn.yaml: -------------------------------------------------------------------------------- 1 | INPUT: 2 | MIN_SIZE_TRAIN: (800,) 3 | MAX_SIZE_TRAIN: 1333 4 | MIN_SIZE_TEST: 800 5 | MAX_SIZE_TEST: 1333 6 | MODEL: 7 | META_ARCHITECTURE: "GeneralizedRCNN" 8 | WEIGHT: "catalog://ImageNetPretrained/MSRA/R-50-GN" 9 | BACKBONE: 10 | CONV_BODY: "R-50-FPN" 11 | RESNETS: # use GN for backbone 12 | BACKBONE_OUT_CHANNELS: 256 13 | STRIDE_IN_1X1: False 14 | TRANS_FUNC: "BottleneckWithGN" 15 | STEM_FUNC: "StemWithGN" 16 | FPN: 17 | USE_GN: True # use GN for FPN 18 | RPN: 19 | USE_FPN: True 20 | ANCHOR_STRIDE: (4, 8, 16, 32, 64) 21 | PRE_NMS_TOP_N_TRAIN: 2000 22 | PRE_NMS_TOP_N_TEST: 1000 23 | POST_NMS_TOP_N_TEST: 1000 24 | FPN_POST_NMS_TOP_N_TEST: 1000 25 | ROI_HEADS: 26 | USE_FPN: True 27 | BATCH_SIZE_PER_IMAGE: 512 28 | POSITIVE_FRACTION: 0.25 29 | ROI_BOX_HEAD: 30 | USE_GN: True # use GN for bbox head 31 | POOLER_RESOLUTION: 7 32 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 33 | POOLER_SAMPLING_RATIO: 2 34 | FEATURE_EXTRACTOR: "FPN2MLPFeatureExtractor" 35 | PREDICTOR: "FPNPredictor" 36 | ROI_MASK_HEAD: 37 | USE_GN: True # use GN for mask head 38 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 39 | CONV_LAYERS: (256, 256, 256, 256) 40 | FEATURE_EXTRACTOR: "MaskRCNNFPNFeatureExtractor" 41 | PREDICTOR: "MaskRCNNC4Predictor" 42 | POOLER_RESOLUTION: 14 43 | POOLER_SAMPLING_RATIO: 2 44 | RESOLUTION: 28 45 | SHARE_BOX_FEATURE_EXTRACTOR: False 46 | MASK_ON: True 47 | DATASETS: 48 | TRAIN: ("coco_2014_train", "coco_2014_valminusminival") 49 | TEST: ("coco_2014_minival",) 50 | DATALOADER: 51 | SIZE_DIVISIBILITY: 32 52 | SOLVER: 53 | # Assume 8 gpus 54 | BASE_LR: 0.02 55 | WEIGHT_DECAY: 0.0001 56 | STEPS: (60000, 80000) 57 | MAX_ITER: 90000 58 | IMS_PER_BATCH: 16 59 | TEST: 60 | IMS_PER_BATCH: 8 61 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/examples/common.h: -------------------------------------------------------------------------------- 1 | /* 2 | Copyright (c) 2013, Philipp Krähenbühl 3 | All rights reserved. 4 | 5 | Redistribution and use in source and binary forms, with or without 6 | modification, are permitted provided that the following conditions are met: 7 | * Redistributions of source code must retain the above copyright 8 | notice, this list of conditions and the following disclaimer. 9 | * Redistributions in binary form must reproduce the above copyright 10 | notice, this list of conditions and the following disclaimer in the 11 | documentation and/or other materials provided with the distribution. 12 | * Neither the name of the Stanford University nor the 13 | names of its contributors may be used to endorse or promote products 14 | derived from this software without specific prior written permission. 15 | 16 | THIS SOFTWARE IS PROVIDED BY Philipp Krähenbühl ''AS IS'' AND ANY 17 | EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED 18 | WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE 19 | DISCLAIMED. IN NO EVENT SHALL Philipp Krähenbühl BE LIABLE FOR ANY 20 | DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES 21 | (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; 22 | LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND 23 | ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT 24 | (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS 25 | SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. 26 | */ 27 | #pragma once 28 | 29 | #include "densecrf.h" 30 | VectorXs getLabeling( const unsigned char * im, int N, int M ); 31 | unsigned char * colorize( const VectorXs & labeling, int W, int H ); 32 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/src/Eigen2Support/Memory.h: -------------------------------------------------------------------------------- 1 | // This file is part of Eigen, a lightweight C++ template library 2 | // for linear algebra. 3 | // 4 | // Copyright (C) 2011 Benoit Jacob 5 | // 6 | // This Source Code Form is subject to the terms of the Mozilla 7 | // Public License v. 2.0. If a copy of the MPL was not distributed 8 | // with this file, You can obtain one at http://mozilla.org/MPL/2.0/. 9 | 10 | #ifndef EIGEN2_MEMORY_H 11 | #define EIGEN2_MEMORY_H 12 | 13 | namespace Eigen { 14 | 15 | inline void* ei_aligned_malloc(size_t size) { return internal::aligned_malloc(size); } 16 | inline void ei_aligned_free(void *ptr) { internal::aligned_free(ptr); } 17 | inline void* ei_aligned_realloc(void *ptr, size_t new_size, size_t old_size) { return internal::aligned_realloc(ptr, new_size, old_size); } 18 | inline void* ei_handmade_aligned_malloc(size_t size) { return internal::handmade_aligned_malloc(size); } 19 | inline void ei_handmade_aligned_free(void *ptr) { internal::handmade_aligned_free(ptr); } 20 | 21 | template inline void* ei_conditional_aligned_malloc(size_t size) 22 | { 23 | return internal::conditional_aligned_malloc(size); 24 | } 25 | template inline void ei_conditional_aligned_free(void *ptr) 26 | { 27 | internal::conditional_aligned_free(ptr); 28 | } 29 | template inline void* ei_conditional_aligned_realloc(void* ptr, size_t new_size, size_t old_size) 30 | { 31 | return internal::conditional_aligned_realloc(ptr, new_size, old_size); 32 | } 33 | 34 | template inline T* ei_aligned_new(size_t size) 35 | { 36 | return internal::aligned_new(size); 37 | } 38 | template inline void ei_aligned_delete(T *ptr, size_t size) 39 | { 40 | return internal::aligned_delete(ptr, size); 41 | } 42 | 43 | } // end namespace Eigen 44 | 45 | #endif // EIGEN2_MACROS_H 46 | -------------------------------------------------------------------------------- /docker/Dockerfile: -------------------------------------------------------------------------------- 1 | ARG CUDA="9.0" 2 | ARG CUDNN="7" 3 | 4 | FROM nvidia/cuda:${CUDA}-cudnn${CUDNN}-devel-ubuntu16.04 5 | 6 | RUN echo 'debconf debconf/frontend select Noninteractive' | debconf-set-selections 7 | 8 | # install basics 9 | RUN apt-get update -y \ 10 | && apt-get install -y apt-utils git curl ca-certificates bzip2 cmake tree htop bmon iotop g++ \ 11 | && apt-get install -y libglib2.0-0 libsm6 libxext6 libxrender-dev 12 | 13 | # Install Miniconda 14 | RUN curl -so /miniconda.sh https://repo.continuum.io/miniconda/Miniconda3-latest-Linux-x86_64.sh \ 15 | && chmod +x /miniconda.sh \ 16 | && /miniconda.sh -b -p /miniconda \ 17 | && rm /miniconda.sh 18 | 19 | ENV PATH=/miniconda/bin:$PATH 20 | 21 | # Create a Python 3.6 environment 22 | RUN /miniconda/bin/conda install -y conda-build \ 23 | && /miniconda/bin/conda create -y --name py36 python=3.6.7 \ 24 | && /miniconda/bin/conda clean -ya 25 | 26 | ENV CONDA_DEFAULT_ENV=py36 27 | ENV CONDA_PREFIX=/miniconda/envs/$CONDA_DEFAULT_ENV 28 | ENV PATH=$CONDA_PREFIX/bin:$PATH 29 | ENV CONDA_AUTO_UPDATE_CONDA=false 30 | 31 | RUN conda install -y ipython 32 | RUN pip install ninja yacs cython matplotlib opencv-python 33 | 34 | # Install PyTorch 1.0 Nightly 35 | ARG CUDA 36 | RUN conda install pytorch-nightly cudatoolkit=${CUDA} -c pytorch \ 37 | && conda clean -ya 38 | 39 | # Install TorchVision master 40 | RUN git clone https://github.com/pytorch/vision.git \ 41 | && cd vision \ 42 | && python setup.py install 43 | 44 | # install pycocotools 45 | RUN git clone https://github.com/cocodataset/cocoapi.git \ 46 | && cd cocoapi/PythonAPI \ 47 | && python setup.py build_ext install 48 | 49 | # install PyTorch Detection 50 | ARG FORCE_CUDA="1" 51 | ENV FORCE_CUDA=${FORCE_CUDA} 52 | RUN git clone https://github.com/facebookresearch/maskrcnn-benchmark.git \ 53 | && cd maskrcnn-benchmark \ 54 | && python setup.py build develop 55 | 56 | WORKDIR /maskrcnn-benchmark 57 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/Geometry: -------------------------------------------------------------------------------- 1 | #ifndef EIGEN_GEOMETRY_MODULE_H 2 | #define EIGEN_GEOMETRY_MODULE_H 3 | 4 | #include "Core" 5 | 6 | #include "src/Core/util/DisableStupidWarnings.h" 7 | 8 | #include "SVD" 9 | #include "LU" 10 | #include 11 | 12 | #ifndef M_PI 13 | #define M_PI 3.14159265358979323846 14 | #endif 15 | 16 | /** \defgroup Geometry_Module Geometry module 17 | * 18 | * 19 | * 20 | * This module provides support for: 21 | * - fixed-size homogeneous transformations 22 | * - translation, scaling, 2D and 3D rotations 23 | * - quaternions 24 | * - \ref MatrixBase::cross() "cross product" 25 | * - \ref MatrixBase::unitOrthogonal() "orthognal vector generation" 26 | * - some linear components: parametrized-lines and hyperplanes 27 | * 28 | * \code 29 | * #include 30 | * \endcode 31 | */ 32 | 33 | #include "src/Geometry/OrthoMethods.h" 34 | #include "src/Geometry/EulerAngles.h" 35 | 36 | #if EIGEN2_SUPPORT_STAGE > STAGE20_RESOLVE_API_CONFLICTS 37 | #include "src/Geometry/Homogeneous.h" 38 | #include "src/Geometry/RotationBase.h" 39 | #include "src/Geometry/Rotation2D.h" 40 | #include "src/Geometry/Quaternion.h" 41 | #include "src/Geometry/AngleAxis.h" 42 | #include "src/Geometry/Transform.h" 43 | #include "src/Geometry/Translation.h" 44 | #include "src/Geometry/Scaling.h" 45 | #include "src/Geometry/Hyperplane.h" 46 | #include "src/Geometry/ParametrizedLine.h" 47 | #include "src/Geometry/AlignedBox.h" 48 | #include "src/Geometry/Umeyama.h" 49 | 50 | #if defined EIGEN_VECTORIZE_SSE 51 | #include "src/Geometry/arch/Geometry_SSE.h" 52 | #endif 53 | #endif 54 | 55 | #ifdef EIGEN2_SUPPORT 56 | #include "src/Eigen2Support/Geometry/All.h" 57 | #endif 58 | 59 | #include "src/Core/util/ReenableStupidWarnings.h" 60 | 61 | #endif // EIGEN_GEOMETRY_MODULE_H 62 | /* vim: set filetype=cpp et sw=2 ts=2 ai: */ 63 | 64 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/include/Eigen/CholmodSupport: -------------------------------------------------------------------------------- 1 | #ifndef EIGEN_CHOLMODSUPPORT_MODULE_H 2 | #define EIGEN_CHOLMODSUPPORT_MODULE_H 3 | 4 | #include "SparseCore" 5 | 6 | #include "src/Core/util/DisableStupidWarnings.h" 7 | 8 | extern "C" { 9 | #include 10 | } 11 | 12 | /** \ingroup Support_modules 13 | * \defgroup CholmodSupport_Module CholmodSupport module 14 | * 15 | * This module provides an interface to the Cholmod library which is part of the suitesparse package. 16 | * It provides the two following main factorization classes: 17 | * - class CholmodSupernodalLLT: a supernodal LLT Cholesky factorization. 18 | * - class CholmodDecomposiiton: a general L(D)LT Cholesky factorization with automatic or explicit runtime selection of the underlying factorization method (supernodal or simplicial). 19 | * 20 | * For the sake of completeness, this module also propose the two following classes: 21 | * - class CholmodSimplicialLLT 22 | * - class CholmodSimplicialLDLT 23 | * Note that these classes does not bring any particular advantage compared to the built-in 24 | * SimplicialLLT and SimplicialLDLT factorization classes. 25 | * 26 | * \code 27 | * #include 28 | * \endcode 29 | * 30 | * In order to use this module, the cholmod headers must be accessible from the include paths, and your binary must be linked to the cholmod library and its dependencies. 31 | * The dependencies depend on how cholmod has been compiled. 32 | * For a cmake based project, you can use our FindCholmod.cmake module to help you in this task. 33 | * 34 | */ 35 | 36 | #include "src/misc/Solve.h" 37 | #include "src/misc/SparseSolve.h" 38 | 39 | #include "src/CholmodSupport/CholmodSupport.h" 40 | 41 | 42 | #include "src/Core/util/ReenableStupidWarnings.h" 43 | 44 | #endif // EIGEN_CHOLMODSUPPORT_MODULE_H 45 | 46 | -------------------------------------------------------------------------------- /configs/gn_baselines/scratch_e2e_mask_rcnn_R_50_FPN_3x_gn.yaml: -------------------------------------------------------------------------------- 1 | INPUT: 2 | MIN_SIZE_TRAIN: (800,) 3 | MAX_SIZE_TRAIN: 1333 4 | MIN_SIZE_TEST: 800 5 | MAX_SIZE_TEST: 1333 6 | MODEL: 7 | META_ARCHITECTURE: "GeneralizedRCNN" 8 | WEIGHT: "" # no pretrained model 9 | BACKBONE: 10 | CONV_BODY: "R-50-FPN" 11 | FREEZE_CONV_BODY_AT: 0 # finetune all layers 12 | RESNETS: # use GN for backbone 13 | BACKBONE_OUT_CHANNELS: 256 14 | STRIDE_IN_1X1: False 15 | TRANS_FUNC: "BottleneckWithGN" 16 | STEM_FUNC: "StemWithGN" 17 | FPN: 18 | USE_GN: True # use GN for FPN 19 | RPN: 20 | USE_FPN: True 21 | ANCHOR_STRIDE: (4, 8, 16, 32, 64) 22 | PRE_NMS_TOP_N_TRAIN: 2000 23 | PRE_NMS_TOP_N_TEST: 1000 24 | POST_NMS_TOP_N_TEST: 1000 25 | FPN_POST_NMS_TOP_N_TEST: 1000 26 | ROI_HEADS: 27 | USE_FPN: True 28 | BATCH_SIZE_PER_IMAGE: 512 29 | POSITIVE_FRACTION: 0.25 30 | ROI_BOX_HEAD: 31 | USE_GN: True # use GN for bbox head 32 | POOLER_RESOLUTION: 7 33 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 34 | POOLER_SAMPLING_RATIO: 2 35 | FEATURE_EXTRACTOR: "FPN2MLPFeatureExtractor" 36 | PREDICTOR: "FPNPredictor" 37 | ROI_MASK_HEAD: 38 | USE_GN: True # use GN for mask head 39 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 40 | CONV_LAYERS: (256, 256, 256, 256) 41 | FEATURE_EXTRACTOR: "MaskRCNNFPNFeatureExtractor" 42 | PREDICTOR: "MaskRCNNC4Predictor" 43 | POOLER_RESOLUTION: 14 44 | POOLER_SAMPLING_RATIO: 2 45 | RESOLUTION: 28 46 | SHARE_BOX_FEATURE_EXTRACTOR: False 47 | MASK_ON: True 48 | DATASETS: 49 | TRAIN: ("coco_2014_train", "coco_2014_valminusminival") 50 | TEST: ("coco_2014_minival",) 51 | DATALOADER: 52 | SIZE_DIVISIBILITY: 32 53 | SOLVER: 54 | # Assume 8 gpus 55 | BASE_LR: 0.02 56 | WEIGHT_DECAY: 0.0001 57 | STEPS: (210000, 250000) 58 | MAX_ITER: 270000 59 | IMS_PER_BATCH: 16 60 | TEST: 61 | IMS_PER_BATCH: 8 62 | -------------------------------------------------------------------------------- /configs/gn_baselines/e2e_mask_rcnn_R_50_FPN_Xconv1fc_1x_gn.yaml: -------------------------------------------------------------------------------- 1 | INPUT: 2 | MIN_SIZE_TRAIN: (800,) 3 | MAX_SIZE_TRAIN: 1333 4 | MIN_SIZE_TEST: 800 5 | MAX_SIZE_TEST: 1333 6 | MODEL: 7 | META_ARCHITECTURE: "GeneralizedRCNN" 8 | WEIGHT: "catalog://ImageNetPretrained/MSRA/R-50-GN" 9 | BACKBONE: 10 | CONV_BODY: "R-50-FPN" 11 | RESNETS: # use GN for backbone 12 | BACKBONE_OUT_CHANNELS: 256 13 | STRIDE_IN_1X1: False 14 | TRANS_FUNC: "BottleneckWithGN" 15 | STEM_FUNC: "StemWithGN" 16 | FPN: 17 | USE_GN: True # use GN for FPN 18 | RPN: 19 | USE_FPN: True 20 | ANCHOR_STRIDE: (4, 8, 16, 32, 64) 21 | PRE_NMS_TOP_N_TRAIN: 2000 22 | PRE_NMS_TOP_N_TEST: 1000 23 | POST_NMS_TOP_N_TEST: 1000 24 | FPN_POST_NMS_TOP_N_TEST: 1000 25 | ROI_HEADS: 26 | USE_FPN: True 27 | BATCH_SIZE_PER_IMAGE: 512 28 | POSITIVE_FRACTION: 0.25 29 | ROI_BOX_HEAD: 30 | USE_GN: True # use GN for bbox head 31 | POOLER_RESOLUTION: 7 32 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 33 | POOLER_SAMPLING_RATIO: 2 34 | CONV_HEAD_DIM: 256 35 | NUM_STACKED_CONVS: 4 36 | FEATURE_EXTRACTOR: "FPNXconv1fcFeatureExtractor" 37 | PREDICTOR: "FPNPredictor" 38 | ROI_MASK_HEAD: 39 | USE_GN: True # use GN for mask head 40 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 41 | CONV_LAYERS: (256, 256, 256, 256) 42 | FEATURE_EXTRACTOR: "MaskRCNNFPNFeatureExtractor" 43 | PREDICTOR: "MaskRCNNC4Predictor" 44 | POOLER_RESOLUTION: 14 45 | POOLER_SAMPLING_RATIO: 2 46 | RESOLUTION: 28 47 | SHARE_BOX_FEATURE_EXTRACTOR: False 48 | MASK_ON: True 49 | DATASETS: 50 | TRAIN: ("coco_2014_train", "coco_2014_valminusminival") 51 | TEST: ("coco_2014_minival",) 52 | DATALOADER: 53 | SIZE_DIVISIBILITY: 32 54 | SOLVER: 55 | # Assume 8 gpus 56 | BASE_LR: 0.02 57 | WEIGHT_DECAY: 0.0001 58 | STEPS: (60000, 80000) 59 | MAX_ITER: 90000 60 | IMS_PER_BATCH: 16 61 | TEST: 62 | IMS_PER_BATCH: 8 63 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/examples/ppm.h: -------------------------------------------------------------------------------- 1 | /* 2 | Copyright (c) 2011, Philipp Krähenbühl 3 | All rights reserved. 4 | 5 | Redistribution and use in source and binary forms, with or without 6 | modification, are permitted provided that the following conditions are met: 7 | * Redistributions of source code must retain the above copyright 8 | notice, this list of conditions and the following disclaimer. 9 | * Redistributions in binary form must reproduce the above copyright 10 | notice, this list of conditions and the following disclaimer in the 11 | documentation and/or other materials provided with the distribution. 12 | * Neither the name of the Stanford University nor the 13 | names of its contributors may be used to endorse or promote products 14 | derived from this software without specific prior written permission. 15 | 16 | THIS SOFTWARE IS PROVIDED BY Philipp Krähenbühl ''AS IS'' AND ANY 17 | EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED 18 | WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE 19 | DISCLAIMED. IN NO EVENT SHALL Philipp Krähenbühl BE LIABLE FOR ANY 20 | DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES 21 | (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; 22 | LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND 23 | ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT 24 | (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS 25 | SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. 26 | */ 27 | 28 | #pragma once 29 | 30 | 31 | void writePGM( const char * filename, int W, int H, const char * data ); 32 | unsigned char * readPPM( const char * filename, int & W, int & H ); 33 | void writePPM( const char * filename, int W, int H, unsigned char * data ); 34 | -------------------------------------------------------------------------------- /configs/gn_baselines/scratch_e2e_mask_rcnn_R_50_FPN_Xconv1fc_3x_gn.yaml: -------------------------------------------------------------------------------- 1 | INPUT: 2 | MIN_SIZE_TRAIN: (800,) 3 | MAX_SIZE_TRAIN: 1333 4 | MIN_SIZE_TEST: 800 5 | MAX_SIZE_TEST: 1333 6 | MODEL: 7 | META_ARCHITECTURE: "GeneralizedRCNN" 8 | WEIGHT: "" # no pretrained model 9 | BACKBONE: 10 | CONV_BODY: "R-50-FPN" 11 | FREEZE_CONV_BODY_AT: 0 # finetune all layers 12 | RESNETS: # use GN for backbone 13 | BACKBONE_OUT_CHANNELS: 256 14 | STRIDE_IN_1X1: False 15 | TRANS_FUNC: "BottleneckWithGN" 16 | STEM_FUNC: "StemWithGN" 17 | FPN: 18 | USE_GN: True # use GN for FPN 19 | RPN: 20 | USE_FPN: True 21 | ANCHOR_STRIDE: (4, 8, 16, 32, 64) 22 | PRE_NMS_TOP_N_TRAIN: 2000 23 | PRE_NMS_TOP_N_TEST: 1000 24 | POST_NMS_TOP_N_TEST: 1000 25 | FPN_POST_NMS_TOP_N_TEST: 1000 26 | ROI_HEADS: 27 | USE_FPN: True 28 | BATCH_SIZE_PER_IMAGE: 512 29 | POSITIVE_FRACTION: 0.25 30 | ROI_BOX_HEAD: 31 | USE_GN: True # use GN for bbox head 32 | POOLER_RESOLUTION: 7 33 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 34 | POOLER_SAMPLING_RATIO: 2 35 | CONV_HEAD_DIM: 256 36 | NUM_STACKED_CONVS: 4 37 | FEATURE_EXTRACTOR: "FPNXconv1fcFeatureExtractor" 38 | PREDICTOR: "FPNPredictor" 39 | ROI_MASK_HEAD: 40 | USE_GN: True # use GN for mask head 41 | POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125) 42 | CONV_LAYERS: (256, 256, 256, 256) 43 | FEATURE_EXTRACTOR: "MaskRCNNFPNFeatureExtractor" 44 | PREDICTOR: "MaskRCNNC4Predictor" 45 | POOLER_RESOLUTION: 14 46 | POOLER_SAMPLING_RATIO: 2 47 | RESOLUTION: 28 48 | SHARE_BOX_FEATURE_EXTRACTOR: False 49 | MASK_ON: True 50 | DATASETS: 51 | TRAIN: ("coco_2014_train", "coco_2014_valminusminival") 52 | TEST: ("coco_2014_minival",) 53 | DATALOADER: 54 | SIZE_DIVISIBILITY: 32 55 | SOLVER: 56 | # Assume 8 gpus 57 | BASE_LR: 0.02 58 | WEIGHT_DECAY: 0.0001 59 | STEPS: (210000, 250000) 60 | MAX_ITER: 270000 61 | IMS_PER_BATCH: 16 62 | TEST: 63 | IMS_PER_BATCH: 8 64 | -------------------------------------------------------------------------------- /matlab/DenseCRF/include/densecrf/external/liblbfgs/config.h.in: -------------------------------------------------------------------------------- 1 | /* config.h.in. Generated from configure.in by autoheader. */ 2 | 3 | /* Define to 1 if you have the header file. */ 4 | #undef HAVE_DLFCN_H 5 | 6 | /* Define to 1 if you have the header file. */ 7 | #undef HAVE_EMMINTRIN_H 8 | 9 | /* Define to 1 if you have the header file. */ 10 | #undef HAVE_INTTYPES_H 11 | 12 | /* Define to 1 if you have the `m' library (-lm). */ 13 | #undef HAVE_LIBM 14 | 15 | /* Define to 1 if you have the header file. */ 16 | #undef HAVE_MEMORY_H 17 | 18 | /* Define to 1 if you have the header file. */ 19 | #undef HAVE_STDINT_H 20 | 21 | /* Define to 1 if you have the header file. */ 22 | #undef HAVE_STDLIB_H 23 | 24 | /* Define to 1 if you have the header file. */ 25 | #undef HAVE_STRINGS_H 26 | 27 | /* Define to 1 if you have the header file. */ 28 | #undef HAVE_STRING_H 29 | 30 | /* Define to 1 if you have the header file. */ 31 | #undef HAVE_SYS_STAT_H 32 | 33 | /* Define to 1 if you have the header file. */ 34 | #undef HAVE_SYS_TYPES_H 35 | 36 | /* Define to 1 if you have the header file. */ 37 | #undef HAVE_UNISTD_H 38 | 39 | /* Define to 1 if you have the header file. */ 40 | #undef HAVE_XMMINTRIN_H 41 | 42 | /* Name of package */ 43 | #undef PACKAGE 44 | 45 | /* Define to the address where bug reports for this package should be sent. */ 46 | #undef PACKAGE_BUGREPORT 47 | 48 | /* Define to the full name of this package. */ 49 | #undef PACKAGE_NAME 50 | 51 | /* Define to the full name and version of this package. */ 52 | #undef PACKAGE_STRING 53 | 54 | /* Define to the one symbol short name of this package. */ 55 | #undef PACKAGE_TARNAME 56 | 57 | /* Define to the version of this package. */ 58 | #undef PACKAGE_VERSION 59 | 60 | /* Define to 1 if you have the ANSI C header files. */ 61 | #undef STDC_HEADERS 62 | 63 | /* Version number of package */ 64 | #undef VERSION 65 | --------------------------------------------------------------------------------