├── .gitignore
├── LICENSE
├── README.md
├── assets
├── optimal_matching_network.png
├── pix2poly_overall_bg_white.png
├── sfo7.png
└── vertex_sequence_detector.png
├── config.py
├── data_preprocess
├── README.md
├── download_mass_roads_dataset.sh
├── inria_to_coco.py
├── mass_roads_clip_shapefile.py
├── mass_roads_clip_tile_vectors.py
├── mass_roads_tiles_to_patches.py
├── mass_roads_world_to_pixel_coords.py
├── spacenet_convert_16bit_to_8bit.py
├── spacenet_to_coco.py
├── spacenet_world_to_pixel_coords.py
└── whu_buildings_to_coco.py
├── datasets
├── __init__.py
├── dataset_inria_coco.py
├── dataset_mass_roads.py
├── dataset_spacenet_coco.py
└── dataset_whu_buildings_coco.py
├── ddp_utils.py
├── engine.py
├── eval
├── __init__.py
├── hisup_eval_utils
│ └── metrics
│ │ ├── angle_eval.py
│ │ ├── cIoU.py
│ │ └── polis.py
└── topdig_eval_utils
│ └── metrics
│ └── topdig_metrics.py
├── evaluate_mass_roads_predictions.py
├── evaluation.py
├── models
└── model.py
├── postprocess_coco_parts.py
├── predict_inria_coco_val_set.py
├── predict_mass_roads_test_set.py
├── predict_spacenet_coco_val_set.py
├── predict_whu_buildings_coco_test_set.py
├── pyrightconfig.json
├── tokenizer.py
├── train_ddp.py
└── utils.py
/.gitignore:
--------------------------------------------------------------------------------
1 | # Directories
2 | __pycache__
3 | data
4 | runs
5 | .vscode
6 | preds*
7 | crowdai_visualizations
8 | scratch
9 |
10 | # Files
11 | *.jpg
12 | *.png
13 | *.tar.xz
14 | *.gif
15 | *.txt
16 |
17 | # Files ignore
18 | !assets/sfo7.png
19 | !assets/pix2poly_overall_bg_white.png
20 | !assets/vertex_sequence_detector.png
21 | !assets/optimal_matching_network.png
22 |
--------------------------------------------------------------------------------
/LICENSE:
--------------------------------------------------------------------------------
1 | MIT License
2 |
3 | Copyright (c) 2025 Yeshwanth Kumar Adimoolam
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 |
--------------------------------------------------------------------------------
/README.md:
--------------------------------------------------------------------------------
1 |
8 |
9 |
10 | [[Project Webpage](https://yeshwanth95.github.io/Pix2Poly)] [[Paper](https://arxiv.org/abs/2412.07899)] [[Video]()]
11 |
12 | ## UPDATES:
13 |
14 | 1. 05.06.2025 - Pretrained checkpoints for Pix2Poly on the various datasets used in the paper are released. See [pretrained checkpoints](#pretrained-checkpoints).
15 | 2. 21.05.2025 - As reported by the authors of the [$P^3$ dataset](https://arxiv.org/abs/2505.15379), Pix2Poly achieves state-of-the-art results for multimodal building vectorization from image and LiDAR data sources.
16 |
17 | ### Abstract:
18 |
19 | Extraction of building footprint polygons from remotely sensed data is essential for several urban understanding tasks such as reconstruction, navigation, and mapping. Despite significant progress in the area, extracting accurate polygonal building footprints remains an open problem. In this paper, we introduce Pix2Poly, an attention-based end-to-end trainable and differentiable deep neural network capable of directly generating explicit high-quality building footprints in a ring graph format. Pix2Poly employs a generative encoder-decoder transformer to produce a sequence of graph vertex tokens whose connectivity information is learned by an optimal matching network. Compared to previous graph learning methods, ours is a truly end-to-end trainable approach that extracts high-quality building footprints and road networks without requiring complicated, computationally intensive raster loss functions and intricate training pipelines. Upon evaluating Pix2Poly on several complex and challenging datasets, we report that Pix2Poly outperforms state-of-the-art methods in several vector shape quality metrics while being an entirely explicit method.
20 |
21 | ### Method
22 |
23 |
24 |
25 |
26 |
27 |
28 |
29 |
7 |