{"id":13627538,"url":"https://github.com/TuSimple/TuSimple-DUC","last_synced_at":"2025-04-17T00:31:51.656Z","repository":{"id":68971851,"uuid":"109775550","full_name":"TuSimple/TuSimple-DUC","owner":"TuSimple","description":"Understanding Convolution for Semantic Segmentation","archived":false,"fork":false,"pushed_at":"2021-10-26T04:02:46.000Z","size":36,"stargazers_count":602,"open_issues_count":6,"forks_count":118,"subscribers_count":25,"default_branch":"master","last_synced_at":"2024-08-01T22:40:33.673Z","etag":null,"topics":["cityscapes","convolution","deep-learning","deep-neural-networks","mxnet","python","semantic-segmentation"],"latest_commit_sha":null,"homepage":"https://arxiv.org/abs/1702.08502","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"apache-2.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/TuSimple.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null}},"created_at":"2017-11-07T02:19:43.000Z","updated_at":"2024-07-26T22:44:58.000Z","dependencies_parsed_at":null,"dependency_job_id":"9549cb64-10ac-4870-881e-44eb0abb8918","html_url":"https://github.com/TuSimple/TuSimple-DUC","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/TuSimple%2FTuSimple-DUC","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/TuSimple%2FTuSimple-DUC/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/TuSimple%2FTuSimple-DUC/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/TuSimple%2FTuSimple-DUC/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/TuSimple","download_url":"https://codeload.github.com/TuSimple/TuSimple-DUC/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":223735011,"owners_count":17194027,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["cityscapes","convolution","deep-learning","deep-neural-networks","mxnet","python","semantic-segmentation"],"created_at":"2024-08-01T22:00:35.338Z","updated_at":"2024-11-08T18:30:36.707Z","avatar_url":"https://github.com/TuSimple.png","language":"Python","readme":"# TuSimple-DUC\n\nby Panqu Wang, Pengfei Chen, Ye Yuan, Ding Liu, Zehua Huang, Xiaodi Hou, and Garrison Cottrell.\n\n## Introduction\n\nThis repository is for [Understanding Convolution for Semantic Segmentation](https://arxiv.org/abs/1702.08502) (WACV 2018), which achieved state-of-the-art result on the CityScapes, PASCAL VOC 2012, and Kitti Road benchmark.\n\n## Requirement\n\nWe tested our code on:\n\nUbuntu 16.04, Python 2.7 with\n\n[MXNet (0.11.0)](https://github.com/TuSimple/mxnet), numpy(1.13.1), cv2(3.2.0), PIL(4.2.1), and cython(0.25.2)\n\n## Usage\n\n1. Clone the repository:\n\n   ```shell\n   git clone git@github.com:TuSimple/TuSimple-DUC.git\n   python setup.py develop --user\n   ```\n\n2. Download the pretrained model from [Google Drive](https://drive.google.com/drive/folders/0B72xLTlRb0SoREhISlhibFZTRmM?resourcekey=0-g2Fr58Nn591bs5ZvZ0Vlwg\u0026usp=sharing).\n\n3. Build MXNet (only tested on the TuSimple version):\n\n   ```shell\n   git clone --recursive git@github.com:TuSimple/mxnet.git\n   vim make/config.mk (we should have USE_CUDA = 1, modify USE_CUDA_PATH, and have USE_CUDNN = 1 to enable GPU usage.)\n   make -j\n   cd python\n   python setup.py develop --user\n   ```\n\n   For more MXNet tutorials, please refer to the [official documentation](https://mxnet.incubator.apache.org/install/index.html).\n\n3. Training:\n\n   ```shell\n   cd train\n   python train_model.py ../configs/train/train_cityscapes.cfg\n   ```\n\n   The paths/dirs in the ``.cfg`` file need to be specified by the user.\n\n4. Testing\n\n   ```\n   cd test\n   python predict_full_image.py ../configs/test/test_full_image.cfg\n   ```\n\n   The paths/dirs in the ``.cfg`` file need to be specified by the user.\n\n5. Results:\n\n   Modify the ``result_dir`` path in the config file to save the label map and visualizations. The expected scores are:\n\n   (single scale testing denotes as 'ss' and multiple scale testing denotes as 'ms')\n\n   - ResNet101-DUC-HDC on CityScapes testset (mIoU): 79.1(ss) / 80.1(ms)\n   - ResNet152-DUC on VOC2012 (mIoU): 83.1(ss)\n\n## Citation\n\nIf you find the repository is useful for your research, please consider citing:\n\n    @article{wang2017understanding,\n      title={Understanding convolution for semantic segmentation},\n      author={Wang, Panqu and Chen, Pengfei and Yuan, Ye and Liu, Ding and Huang, Zehua and Hou, Xiaodi and Cottrell, Garrison},\n      journal={arXiv preprint arXiv:1702.08502},\n      year={2017}\n    }\n\n## Questions\n\nPlease contact panqu.wang@tusimple.ai or pengfei.chen@tusimple.ai .\n","funding_links":[],"categories":["\u003ca name=\"Vision\"\u003e\u003c/a\u003e2. Vision"],"sub_categories":["2.3 Image Segmentation"],"project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2FTuSimple%2FTuSimple-DUC","html_url":"https://awesome.ecosyste.ms/projects/github.com%2FTuSimple%2FTuSimple-DUC","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2FTuSimple%2FTuSimple-DUC/lists"}