{"id":17154763,"url":"https://github.com/liruilong940607/pose2seg","last_synced_at":"2025-04-04T21:08:19.220Z","repository":{"id":37601861,"uuid":"177575086","full_name":"liruilong940607/Pose2Seg","owner":"liruilong940607","description":"Code for the paper \"Pose2Seg: Detection Free Human Instance Segmentation\" @ CVPR2019.","archived":false,"fork":false,"pushed_at":"2023-09-23T16:02:50.000Z","size":2016,"stargazers_count":532,"open_issues_count":25,"forks_count":135,"subscribers_count":20,"default_branch":"master","last_synced_at":"2025-03-28T20:07:15.820Z","etag":null,"topics":["cvpr2019","human","pose-estimation","segmentation"],"latest_commit_sha":null,"homepage":"http://www.liruilong.cn/projects/pose2seg/index.html","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/liruilong940607.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,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2019-03-25T11:41:29.000Z","updated_at":"2025-03-10T02:32:29.000Z","dependencies_parsed_at":"2024-10-30T09:22:07.964Z","dependency_job_id":null,"html_url":"https://github.com/liruilong940607/Pose2Seg","commit_stats":null,"previous_names":[],"tags_count":1,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/liruilong940607%2FPose2Seg","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/liruilong940607%2FPose2Seg/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/liruilong940607%2FPose2Seg/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/liruilong940607%2FPose2Seg/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/liruilong940607","download_url":"https://codeload.github.com/liruilong940607/Pose2Seg/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":247249525,"owners_count":20908212,"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":["cvpr2019","human","pose-estimation","segmentation"],"created_at":"2024-10-14T21:49:55.490Z","updated_at":"2025-04-04T21:08:19.194Z","avatar_url":"https://github.com/liruilong940607.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Pose2Seg\n\n*Official* code for the paper \"Pose2Seg: Detection Free Human Instance Segmentation\"[[ProjectPage]](http://www.liruilong.cn/projects/pose2seg/index.html)[[arXiv]](https://arxiv.org/abs/1803.10683) @ CVPR2019.\n\nThe *OCHuman dataset* proposed in our paper is released [here](https://github.com/liruilong940607/OCHumanApi)\n\n\u003cdiv align=\"center\"\u003e\n\u003cimg src=\"figures/pipeline.jpg\" width=\"1000px\"/\u003e\n\u003cp\u003e Pipeline of our pose-based instance segmentation framework.\u003c/p\u003e\n\u003c/div\u003e\n\n## Setup environment\n\n``` bash\npip install cython matplotlib tqdm opencv-python scipy pyyaml numpy\npip install torchvision torch\n\ncd ~/github-public/cocoapi/PythonAPI/\npython setup.py build_ext install\ncd -\n```\n\n## Download data\n\n- COCO 2017\n    - [COCO 2017 Train images [118K/18GB]](http://images.cocodataset.org/zips/train2017.zip)\n    - [COCO 2017 Val images [5K/1GB]](http://images.cocodataset.org/zips/val2017.zip)\n    - [COCOPersons Train Annotation (person_keypoints_train2017_pose2seg.json) [166MB]](https://github.com/liruilong940607/Pose2Seg/releases/download/data/person_keypoints_train2017_pose2seg.json)\n    - [COCOPersons Val Annotation (person_keypoints_val2017_pose2seg.json) [7MB]](https://github.com/liruilong940607/Pose2Seg/releases/download/data/person_keypoints_val2017_pose2seg.json)\n    \n- OCHuman\n    - [images [667MB] \u0026 annotations](https://cg.cs.tsinghua.edu.cn/dataset/form.html?dataset=ochuman)\n    \n**Note**: \n`person_keypoints_(train/val)2017_pose2seg.json` is a subset of `person_keypoints_(train/val)2017.json` (in [COCO2017 Train/Val annotations](http://images.cocodataset.org/annotations/annotations_trainval2017.zip)). We choose those instances with both keypoint and segmentation annotations for our experiments.\n\n## Setup data\n\nThe `data` folder should be like this:\n\n    data  \n    ├── coco2017\n    │   ├── annotations  \n    │   │   ├── person_keypoints_train2017_pose2seg.json \n    │   │   ├── person_keypoints_val2017_pose2seg.json \n    │   ├── train2017  \n    │   │   ├── ####.jpg  \n    │   ├── val2017  \n    │   │   ├── ####.jpg  \n    ├── OCHuman \n    │   ├── annotations  \n    │   │   ├── ochuman_coco_format_test_range_0.00_1.00.json   \n    │   │   ├── ochuman_coco_format_val_range_0.00_1.00.json   \n    │   ├── images  \n    │   │   ├── ####.jpg \n\n## How to train\n\n``` bash\npython train.py\n```\n\n**Note**: Currently we only support for single-gpu training.\n\n## How to test\n\nThis allows you to test the model on (1) COCOPersons val set and (2) OCHuman val \u0026 test set.\n\n``` bash\npython test.py --weights last.pkl --coco --OCHuman\n```\n\nWe retrained our model using this repo, and got similar results with our paper. The final weights can be download [here](https://drive.google.com/file/d/193i8b40MJFxawcJoNLq1sG0vhAeLoVJG/view?usp=sharing).\n\n## About Human Pose Templates in COCO\n\n\u003cdiv align=\"center\"\u003e\n\u003cimg src=\"figures/pose_templates.png\" width=\"500px\"/\u003e\n\u003cp\u003e Pose templates clustered using K-means on COCO.\u003c/p\u003e\n\u003c/div\u003e\n\nThis repo already contains a template file `modeling/templates.json` which was used in our paper. But you are free to explore different cluster parameters as discussed in our paper. See [visualize_cluster.ipynb](visualize_cluster.ipynb) for an example.\n\n\n\n\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fliruilong940607%2Fpose2seg","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fliruilong940607%2Fpose2seg","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fliruilong940607%2Fpose2seg/lists"}