├── .gitignore ├── LICENSE ├── README.md ├── dataset ├── __init__.py ├── build.py ├── datasets.py ├── functional.py ├── loader.py ├── masking_generator.py ├── pretrain_datasets.py ├── rand_augment.py ├── random_erasing.py ├── transforms.py ├── video_transforms.py └── volume_transforms.py ├── docs ├── DATASET.md ├── FINETUNE.md ├── INSTALL.md ├── MODEL_ZOO.md ├── PRETRAIN.md └── TAD.md ├── engine_for_finetuning.py ├── engine_for_pretraining.py ├── extract_tad_feature.py ├── misc ├── VideoMAEv2_flowchart.png ├── k710_identical_label_merge.json ├── label_710to400.json ├── label_710to600.json ├── label_710to700.json ├── label_map_k400.txt ├── label_map_k600.txt ├── label_map_k700.txt └── label_map_k710.txt ├── models ├── __init__.py ├── modeling_finetune.py └── modeling_pretrain.py ├── optim_factory.py ├── requirements.txt ├── run_class_finetuning.py ├── run_mae_pretraining.py ├── scripts ├── finetune │ ├── README.md │ ├── vit_b_k400_ft.sh │ ├── vit_b_k710_ft.sh │ ├── vit_b_ssv2_ft.sh │ ├── vit_g_k400_ft.sh │ ├── vit_g_k710_ft.sh │ ├── vit_g_k710_it_hmdb51_ft.sh │ ├── vit_g_k710_it_k400_ft.sh │ ├── vit_g_k710_it_k600_ft.sh │ ├── vit_g_k710_it_ucf101_ft.sh │ ├── vit_g_ssv2_ft.sh │ ├── vit_h_k400_ft.sh │ ├── vit_h_k600_ft.sh │ ├── vit_h_k710_ft.sh │ ├── vit_h_ssv2_ft.sh │ ├── vit_l_k400_ft.sh │ └── vit_l_ssv2_ft.sh └── pretrain │ ├── README.md │ ├── vit_b_hybrid_pt.sh │ ├── vit_g_hybrid_pt.sh │ ├── vit_h_hybrid_pt.sh │ └── vit_l_hybrid_pt.sh └── utils.py /.gitignore: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/.gitignore -------------------------------------------------------------------------------- /LICENSE: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/LICENSE -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/README.md -------------------------------------------------------------------------------- /dataset/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/dataset/__init__.py -------------------------------------------------------------------------------- /dataset/build.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/dataset/build.py -------------------------------------------------------------------------------- /dataset/datasets.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/dataset/datasets.py -------------------------------------------------------------------------------- /dataset/functional.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/dataset/functional.py -------------------------------------------------------------------------------- /dataset/loader.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/dataset/loader.py -------------------------------------------------------------------------------- /dataset/masking_generator.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/dataset/masking_generator.py -------------------------------------------------------------------------------- /dataset/pretrain_datasets.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/dataset/pretrain_datasets.py -------------------------------------------------------------------------------- /dataset/rand_augment.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/dataset/rand_augment.py -------------------------------------------------------------------------------- /dataset/random_erasing.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/dataset/random_erasing.py -------------------------------------------------------------------------------- /dataset/transforms.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/dataset/transforms.py -------------------------------------------------------------------------------- /dataset/video_transforms.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/dataset/video_transforms.py -------------------------------------------------------------------------------- /dataset/volume_transforms.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/dataset/volume_transforms.py -------------------------------------------------------------------------------- /docs/DATASET.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/docs/DATASET.md -------------------------------------------------------------------------------- /docs/FINETUNE.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/docs/FINETUNE.md -------------------------------------------------------------------------------- /docs/INSTALL.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/docs/INSTALL.md -------------------------------------------------------------------------------- /docs/MODEL_ZOO.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/docs/MODEL_ZOO.md -------------------------------------------------------------------------------- /docs/PRETRAIN.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/docs/PRETRAIN.md -------------------------------------------------------------------------------- /docs/TAD.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/docs/TAD.md -------------------------------------------------------------------------------- /engine_for_finetuning.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/engine_for_finetuning.py -------------------------------------------------------------------------------- /engine_for_pretraining.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/engine_for_pretraining.py -------------------------------------------------------------------------------- /extract_tad_feature.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/extract_tad_feature.py -------------------------------------------------------------------------------- /misc/VideoMAEv2_flowchart.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/misc/VideoMAEv2_flowchart.png -------------------------------------------------------------------------------- /misc/k710_identical_label_merge.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/misc/k710_identical_label_merge.json -------------------------------------------------------------------------------- /misc/label_710to400.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/misc/label_710to400.json -------------------------------------------------------------------------------- /misc/label_710to600.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/misc/label_710to600.json -------------------------------------------------------------------------------- /misc/label_710to700.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/misc/label_710to700.json -------------------------------------------------------------------------------- /misc/label_map_k400.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/misc/label_map_k400.txt -------------------------------------------------------------------------------- /misc/label_map_k600.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/misc/label_map_k600.txt -------------------------------------------------------------------------------- /misc/label_map_k700.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/misc/label_map_k700.txt -------------------------------------------------------------------------------- /misc/label_map_k710.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/misc/label_map_k710.txt -------------------------------------------------------------------------------- /models/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/models/__init__.py -------------------------------------------------------------------------------- /models/modeling_finetune.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/models/modeling_finetune.py -------------------------------------------------------------------------------- /models/modeling_pretrain.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/models/modeling_pretrain.py -------------------------------------------------------------------------------- /optim_factory.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/optim_factory.py -------------------------------------------------------------------------------- /requirements.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/requirements.txt -------------------------------------------------------------------------------- /run_class_finetuning.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/run_class_finetuning.py -------------------------------------------------------------------------------- /run_mae_pretraining.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/run_mae_pretraining.py -------------------------------------------------------------------------------- /scripts/finetune/README.md: -------------------------------------------------------------------------------- 1 | ../../docs/FINETUNE.md -------------------------------------------------------------------------------- /scripts/finetune/vit_b_k400_ft.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/scripts/finetune/vit_b_k400_ft.sh -------------------------------------------------------------------------------- /scripts/finetune/vit_b_k710_ft.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/scripts/finetune/vit_b_k710_ft.sh -------------------------------------------------------------------------------- /scripts/finetune/vit_b_ssv2_ft.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/scripts/finetune/vit_b_ssv2_ft.sh -------------------------------------------------------------------------------- /scripts/finetune/vit_g_k400_ft.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/scripts/finetune/vit_g_k400_ft.sh -------------------------------------------------------------------------------- /scripts/finetune/vit_g_k710_ft.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/scripts/finetune/vit_g_k710_ft.sh -------------------------------------------------------------------------------- /scripts/finetune/vit_g_k710_it_hmdb51_ft.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/scripts/finetune/vit_g_k710_it_hmdb51_ft.sh -------------------------------------------------------------------------------- /scripts/finetune/vit_g_k710_it_k400_ft.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/scripts/finetune/vit_g_k710_it_k400_ft.sh -------------------------------------------------------------------------------- /scripts/finetune/vit_g_k710_it_k600_ft.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/scripts/finetune/vit_g_k710_it_k600_ft.sh -------------------------------------------------------------------------------- /scripts/finetune/vit_g_k710_it_ucf101_ft.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/scripts/finetune/vit_g_k710_it_ucf101_ft.sh -------------------------------------------------------------------------------- /scripts/finetune/vit_g_ssv2_ft.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/scripts/finetune/vit_g_ssv2_ft.sh -------------------------------------------------------------------------------- /scripts/finetune/vit_h_k400_ft.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/scripts/finetune/vit_h_k400_ft.sh -------------------------------------------------------------------------------- /scripts/finetune/vit_h_k600_ft.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/scripts/finetune/vit_h_k600_ft.sh -------------------------------------------------------------------------------- /scripts/finetune/vit_h_k710_ft.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/scripts/finetune/vit_h_k710_ft.sh -------------------------------------------------------------------------------- /scripts/finetune/vit_h_ssv2_ft.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/scripts/finetune/vit_h_ssv2_ft.sh -------------------------------------------------------------------------------- /scripts/finetune/vit_l_k400_ft.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/scripts/finetune/vit_l_k400_ft.sh -------------------------------------------------------------------------------- /scripts/finetune/vit_l_ssv2_ft.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/scripts/finetune/vit_l_ssv2_ft.sh -------------------------------------------------------------------------------- /scripts/pretrain/README.md: -------------------------------------------------------------------------------- 1 | ../../docs/PRETRAIN.md -------------------------------------------------------------------------------- /scripts/pretrain/vit_b_hybrid_pt.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/scripts/pretrain/vit_b_hybrid_pt.sh -------------------------------------------------------------------------------- /scripts/pretrain/vit_g_hybrid_pt.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/scripts/pretrain/vit_g_hybrid_pt.sh -------------------------------------------------------------------------------- /scripts/pretrain/vit_h_hybrid_pt.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/scripts/pretrain/vit_h_hybrid_pt.sh -------------------------------------------------------------------------------- /scripts/pretrain/vit_l_hybrid_pt.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/scripts/pretrain/vit_l_hybrid_pt.sh -------------------------------------------------------------------------------- /utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/OpenGVLab/VideoMAEv2/HEAD/utils.py --------------------------------------------------------------------------------