{"id":23774447,"url":"https://github.com/jeffffffli/res-loglikelihood-regression","last_synced_at":"2025-04-06T16:15:40.846Z","repository":{"id":45951816,"uuid":"388742230","full_name":"jeffffffli/res-loglikelihood-regression","owner":"jeffffffli","description":"Code for \"Human Pose Regression with Residual Log-likelihood Estimation\", ICCV 2021 Oral","archived":false,"fork":false,"pushed_at":"2021-11-25T03:47:29.000Z","size":66227,"stargazers_count":429,"open_issues_count":33,"forks_count":44,"subscribers_count":6,"default_branch":"master","last_synced_at":"2025-03-30T15:12:15.812Z","etag":null,"topics":["2d-human-pose","3d-human-pose","human-pose-estimation","iccv","iccv2021","pytorch","regression"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/jeffffffli.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null}},"created_at":"2021-07-23T09:07:34.000Z","updated_at":"2025-03-26T08:36:31.000Z","dependencies_parsed_at":"2022-07-18T17:00:33.312Z","dependency_job_id":null,"html_url":"https://github.com/jeffffffli/res-loglikelihood-regression","commit_stats":null,"previous_names":["jeffffffli/res-loglikelihood-regression","jeff-sjtu/res-loglikelihood-regression"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/jeffffffli%2Fres-loglikelihood-regression","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/jeffffffli%2Fres-loglikelihood-regression/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/jeffffffli%2Fres-loglikelihood-regression/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/jeffffffli%2Fres-loglikelihood-regression/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/jeffffffli","download_url":"https://codeload.github.com/jeffffffli/res-loglikelihood-regression/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":247509238,"owners_count":20950232,"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":["2d-human-pose","3d-human-pose","human-pose-estimation","iccv","iccv2021","pytorch","regression"],"created_at":"2025-01-01T06:08:54.723Z","updated_at":"2025-04-06T16:15:40.826Z","avatar_url":"https://github.com/jeffffffli.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Human Pose Regression with Residual Log-likelihood Estimation\n\n[[`Paper`](https://jeffli.site/res-loglikelihood-regression/resources/ICCV21-RLE.pdf)]\n[[`arXiv`](https://arxiv.org/abs/2107.11291)]\n[[`Project Page`](https://jeffli.site/res-loglikelihood-regression/)]\n\n\u003e [Human Pose Regression with Residual Log-likelihood Estimation](https://jeffli.site/res-loglikelihood-regression/resources/ICCV21-RLE.pdf)  \n\u003e Jiefeng Li, Siyuan Bian, Ailing Zeng, Can Wang, Bo Pang, Wentao Liu, Cewu Lu  \n\u003e ICCV 2021 Oral  \n\n\u003cdiv align=\"center\"\u003e\n    \u003cimg src=\"assets/rle.jpg\", width=\"600\" alt\u003e\u003cbr\u003e\n    Regression with Residual Log-likelihood Estimation\n\u003c/div\u003e\n\n## TODO\n- [ ] Provide minimal implementation of RLE loss.\n- [ ] Add model zoo.\n- [x] Provide implementation on Human3.6M dataset.\n- [x] Provide implementation on COCO dataset.\n\n### Installation\n1. Install pytorch \u003e= 1.1.0 following official instruction.\n2. Install `rlepose`:\n``` bash\npip install cython\npython setup.py develop\n```\n3. Install [COCOAPI](https://github.com/cocodataset/cocoapi).\n``` bash\npip install -U 'git+https://github.com/cocodataset/cocoapi.git#subdirectory=PythonAPI'\n```\n4. Init `data` directory:\n``` bash\nmkdir data\n```\n5. Download [COCO](https://cocodataset.org/#download) data, [MPII](http://human-pose.mpi-inf.mpg.de/#download) ([annotations](https://drive.google.com/file/d/1--EQZnCJI_XJIc9_bw-dzw3MrRFLMptw/view?usp=sharing)) and [Human3.6M](https://drive.google.com/drive/folders/1kgVH-GugrLoc9XyvP6nRoaFpw3TmM5xK) data (from [PoseNet](https://github.com/mks0601/3DMPPE_POSENET_RELEASE) or [ours](https://drive.google.com/drive/folders/1sF2xjAfvEw7rvNaQJqadAU2QDUVFfhQH?usp=sharing)):\n```\n|-- data\n`-- |-- coco\n    |   |-- annotations\n    |   |   |-- person_keypoints_train2017.json\n    |   |   `-- person_keypoints_val2017.json\n    |   `-- images\n    |       |-- train2017\n    |       |   |-- 000000000009.jpg\n    |       |   |-- 000000000025.jpg\n    |       |   |-- 000000000030.jpg\n    |       |   |-- ... \n    |       `-- val2017\n    |           |-- 000000000139.jpg\n    |           |-- 000000000285.jpg\n    |           |-- 000000000632.jpg\n    |           |-- ... \n    |-- mpii\n    |   |-- annotations\n    |   |   `-- annot_mpii.json\n    |   `-- images\n            |-- 000001163.jpg\n            |-- 000003072.jpg\n            |-- 000004812.jpg\n            |--- ...\n    |-- h36m\n    `-- |-- annotations\n        |   |-- Sample_trainmin_train_Human36M_protocol_2.json\n        |   `-- Sample_64_test_Human36M_protocol_2.json\n        `-- images\n            |-- s_01_act_02_subact_01_ca_01\n            |   |-- ...\n            |-- s_01_act_02_subact_01_ca_02\n            |   |-- ...\n            `-- ... \n```\n## Training\n\n### Train on MSCOCO\n``` bash\n./scripts/train.sh ./configs/256x192_res50_regress-flow.yaml train_rle_coco\n```\n\n### Train on Human3.6M\n``` bash\n./scripts/train.sh ./configs/256x192_res50_3d_h36mmpii-flow.yaml train_rle_h36m\n```\n\n## Evaluation\n\n### Validate on MSCOCO\nDownload the pretrained model from [Google Drive](https://drive.google.com/file/d/1YBHqNKkxIVv8CqgDxkezC-4vyKpx-zXK/view?usp=sharing).\n``` bash\n./scripts/validate.sh ./configs/256x192_res50_regress-flow.yaml ./coco-laplace-rle.pth\n```\n\n### Validate on Human3.6M\nDownload the pretrained model from [Google Drive](https://drive.google.com/file/d/1v2ZhembnFyJ_FXGHEOCzGaM-tAVFMy7A/view?usp=sharing).\n``` bash\n./scripts/validate.sh ./configs/256x192_res50_3d_h36mmpii-flow.yaml ./h36m-laplace-rle.pth\n```\n\n### Citing\nIf our code helps your research, please consider citing the following paper:\n```\n@inproceedings{li2021human,\n    title={Human Pose Regression with Residual Log-likelihood Estimation},\n    author={Li, Jiefeng and Bian, Siyuan and Zeng, Ailing and Wang, Can and Pang, Bo and Liu, Wentao and Lu, Cewu},\n    booktitle={ICCV},\n    year={2021}\n}\n```","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fjeffffffli%2Fres-loglikelihood-regression","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fjeffffffli%2Fres-loglikelihood-regression","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fjeffffffli%2Fres-loglikelihood-regression/lists"}