{"id":20503152,"url":"https://github.com/idea-research/click-pose","last_synced_at":"2025-07-30T11:38:34.697Z","repository":{"id":190332863,"uuid":"681175535","full_name":"IDEA-Research/Click-Pose","owner":"IDEA-Research","description":"[ICCV 2023] Official implementation of the paper \"Neural Interactive Keypoint Detection\"","archived":false,"fork":false,"pushed_at":"2023-10-12T08:29:50.000Z","size":69722,"stargazers_count":80,"open_issues_count":2,"forks_count":3,"subscribers_count":3,"default_branch":"master","last_synced_at":"2025-04-13T20:16:33.089Z","etag":null,"topics":["annotation-tool","human-in-the-loop","iccv2023","pose-estimation"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"other","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/IDEA-Research.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":"2023-08-21T12:40:18.000Z","updated_at":"2025-03-13T15:17:40.000Z","dependencies_parsed_at":null,"dependency_job_id":"3745eb26-f01a-4038-8dd4-734470a36e19","html_url":"https://github.com/IDEA-Research/Click-Pose","commit_stats":null,"previous_names":["idea-research/click-pose"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/IDEA-Research/Click-Pose","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/IDEA-Research%2FClick-Pose","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/IDEA-Research%2FClick-Pose/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/IDEA-Research%2FClick-Pose/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/IDEA-Research%2FClick-Pose/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/IDEA-Research","download_url":"https://codeload.github.com/IDEA-Research/Click-Pose/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/IDEA-Research%2FClick-Pose/sbom","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":267858828,"owners_count":24155972,"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","status":"online","status_checked_at":"2025-07-30T02:00:09.044Z","response_time":70,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"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":["annotation-tool","human-in-the-loop","iccv2023","pose-estimation"],"created_at":"2024-11-15T19:29:31.276Z","updated_at":"2025-07-30T11:38:34.668Z","avatar_url":"https://github.com/IDEA-Research.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Neural Interactive Keypoint Detection\nThis is the official pytorch implementation of our ICCV 2023 paper \"[Neural Interactive Keypoint Detection](https://arxiv.org/pdf/2308.10174.pdf).\" \n\n[Jie Yang](https://github.com/yangjie-cv), [Ailing Zeng](https://ailingzeng.site/), [Feng Li](https://scholar.google.com/citations?user=ybRe9GcAAAAJ\u0026hl=zh-CN), [Shilong Liu](http://www.lsl.zone/), [Ruimao Zhang](http://www.zhangruimao.site/), [Lei Zhang](https://www.leizhang.org/)\n\n**Keywords**: 👯 Multi-person 2D pose estimation, 💃 Human-in-the-loop, 🤝Interactive model \n\n## ❤️ Highlights\n- Click-Pose has been supported in our **[DeepDataSpace](https://github.com/IDEA-Research/deepdataspace)** platform. See details for *How to perform intelligent labeling with DDS [here](https://docs.deepdataspace.com/tutorials/ai-annotation.html).*\n- All models for COCO, Human-Art, OCHuman, and CrowdPose are released! \n- **Work flow**: 🤖 Model localizes all keypoints -\u003e 👨 User corrects a few wrong keypoints -\u003e 🤖 Model refines other keypoints\n\n\u003cimg src=\"assets/dds_kpt.gif\" /\u003e  \n\n## 💙 Click-Pose\n\n- 👇 We first propose an **interactive keypoint detection task** for efficient keypoint annotation.\n\u003cimg src=\"assets/main_clickpose.jpg\" /\u003e\n\n- 👇 We present the first neural interactive keypoint detection framework, Click-Pose, an end-to-end baseline to annotate multi-person 2D keypoints given an image. \n\u003cimg src=\"assets/framework_clickpose.jpg\" /\u003e  \n\n- 👇 Click-Pose is more than **10** times faster than manual annotation. Importantly, it significantly alleviates model bias in out-of-domain annotation (e.g., on Human-Art), reducing the time required by **83%** compared to state-of-the-art model annotation ([ViTPose](https://github.com/ViTAE-Transformer/ViTPose)) with manual correction.\n\u003cimg src=\"assets/cost.png\" style=\"height:300px\"/\u003e\n\n\n\n\n## 🚀 Model Zoo\n\n### 1. Model-Only Results \n\n#### COCO val2017 set\n\n|   Model    | Backbone  | Lr schd | mAP  | AP\u003csup\u003e50\u003c/sup\u003e | AP\u003csup\u003e75\u003c/sup\u003e | AP\u003csup\u003eM\u003c/sup\u003e | AP\u003csup\u003eL\u003c/sup\u003e | Time (ms) |                                                 Model                                                 |\n|:----------:|:---------:|:-------:|:----:|:---------------:|:---------------:|:--------------:|:--------------:|:---------:|:-----------------------------------------------------------------------------------------------------:|\n|  ED-Pose   | ResNet-50 |   60e   | 71.7 |      89.7       |      78.8       |      66.2      |      79.7      |    51     |                          [GitHub](https://github.com/IDEA-Research/ED-Pose), [Model](https://drive.google.com/file/d/1Q5OpZeCvaSgqC0NlKeRiJFmHBtusxnjX/view?usp=sharing)                        |\n| Click-Pose |   ResNet-50    |   40e   | 73.0 |      90.4       |      80.0       |      68.1      |      80.5      |    48     | [Google Drive](https://drive.google.com/file/d/1_rp12m0fkpSc7LQ1oXeifdt8SbwcSHtS/view?usp=sharing) |\n\n#### Human-Art val set\n\n|   Model    |   Backbone     | mAP  | AP\u003csup\u003eM\u003c/sup\u003e | AP\u003csup\u003eL\u003c/sup\u003e |                                           Model                                                |\n|:----------:|:-------------:|:----:|:--------------:|:--------------:|:-----------------------------------------------------------------------------------------------------:|\n|  ED-Pose   |     ResNet-50        | 37.5 |      7.6       |      41.1      |    [GitHub](https://github.com/IDEA-Research/ED-Pose), [Model](https://drive.google.com/file/d/1Q5OpZeCvaSgqC0NlKeRiJFmHBtusxnjX/view?usp=sharing)          |\n| Click-Pose |    ResNet-50       | 40.5 |      8.3       |      44.2      | [Google Drive](https://drive.google.com/file/d/1_rp12m0fkpSc7LQ1oXeifdt8SbwcSHtS/view?usp=sharing) |\n\n#### OCHuman test set\n\n|   Model    |   Backbone     | mAP  | AP\u003csup\u003e50\u003c/sup\u003e | AP\u003csup\u003e75\u003c/sup\u003e |                                           Model                                                |\n|:----------:|:-------------:|:----:|:---------------:|:---------------:|:-----------------------------------------------------------------------------------------------------:|\n|  ED-Pose   |     ResNet-50        | 31.4 |      39.5       |      35.1       |    [GitHub](https://github.com/IDEA-Research/ED-Pose), [Model](https://drive.google.com/file/d/1Q5OpZeCvaSgqC0NlKeRiJFmHBtusxnjX/view?usp=sharing)          |\n| Click-Pose |    ResNet-50       | 33.9 |      43.4       |      37.5       | [Google Drive](https://drive.google.com/file/d/1_rp12m0fkpSc7LQ1oXeifdt8SbwcSHtS/view?usp=sharing) |\n\nNote that the model is trained on COCO train2017 set and tested on COCO val2017 set, Human-Art val set, and OCHuman test set.\n\n### 2. Neural Interactive  Results \n\n#### In-domain Annotation (COCO val2017)\n\n|   Model    |   Backbone     | NoC@85 | NoC@90 | NoC@95 |                                           Model                                                |\n|:----------:|:-------------:|:------:|:------:|:------:|:-----------------------------------------------------------------------------------------------------:|\n|  ViTPose   |     ViT-Huge        |  1.46  |  2.15  |  2.87  |     [GitHub](https://github.com/ViTAE-Transformer/ViTPose), [Model](https://1drv.ms/u/s!AimBgYV7JjTlgShLMI-kkmvNfF_h?e=dEhGHe)      |\n| Click-Pose |    ResNet-50       |  0.95  |  1.48  |  1.97  | [Google Drive](https://drive.google.com/file/d/184RIVxFVrDho4Nw5Yquh6fedTKpsZVYX/view?usp=sharing) |\n\n\n\n#### Out-of-domain Annotation (Human-Art val)\n\n|   Model    |   Backbone     | NoC@85 | NoC@90 | NoC@95 |                                           Model                                                |\n|:----------:|:-------------:|:------:|:------:|:------:|:-----------------------------------------------------------------------------------------------------:|\n|  ViTPose   |     ViT-Huge        |  9.12  |  9.79  | 10.13  |     [GitHub](https://github.com/ViTAE-Transformer/ViTPose), [Model](https://1drv.ms/u/s!AimBgYV7JjTlgShLMI-kkmvNfF_h?e=dEhGHe)     |\n| Click-Pose |    ResNet-50       |  4.82  |  5.81  |  6.45  | [Google Drive](https://drive.google.com/file/d/184RIVxFVrDho4Nw5Yquh6fedTKpsZVYX/view?usp=sharing) |\n\n\n## 🔨 Environment Setup \n\n\u003cdetails\u003e\n  \u003csummary\u003eInstallation\u003c/summary\u003e\n  \n  We use the [ED-Pose](https://github.com/IDEA-Research/ED-Pose) as our codebase. We test our models under ```python=3.7.3,pytorch=1.9.0,cuda=11.1```. Other versions might be available as well.\n\n   1. Clone this repo\n   ```sh\n   git clone https://github.com/IDEA-Research/Click-Pose.git\n   cd Click-Pose\n   ```\n\n   2. Install Pytorch and torchvision\n\n   Follow the instruction on https://pytorch.org/get-started/locally/.\n   ```sh\n   # an example:\n   conda install -c pytorch pytorch torchvision\n   ```\n\n   3. Install other needed packages\n   ```sh\n   pip install -r requirements.txt\n   ```\n\n   4. Compiling CUDA operators\n   ```sh\n   cd models/clickpose/ops\n   python setup.py build install\n   # unit test (should see all checking is True)\n   python test.py\n   cd ../../..\n   ```\n\u003c/details\u003e\n\n\u003cdetails\u003e\n  \u003csummary\u003eData Preparation\u003c/summary\u003e\n\n**For COCO data**, please download from [COCO download](http://cocodataset.org/#download). \nThe coco_dir should look like this:\n```\n|-- Click-Pose\n`-- |-- coco_dir\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```\n**For Human-Art data**, please download from [Human-Art download](https://github.com/IDEA-Research/HumanArt), \nThe humanart_dir should look like this:\n```\n|-- Click-Pose\n`-- |-- humanart_dir\n    `-- |-- annotations \n        |   |-- training_humanart.json\n        |   |-- validation_humanart.json\n        `-- images\n            |-- 2D_virtual_human\n                |-- ...\n            |-- 3D_virtual_human\n                |-- ...\n            |-- real_human\n                |-- ...\n```\n\n\n**For CrowdPose data**, please download from [CrowdPose download](https://github.com/Jeff-sjtu/CrowdPose#dataset), \nThe crowdpose_dir should look like this:\n```\n|-- Click-Pose\n`-- |-- crowdpose_dir\n    `-- |-- json\n        |   |-- crowdpose_train.json\n        |   |-- crowdpose_val.json\n        |   |-- crowdpose_trainval.json (generated by util/crowdpose_concat_train_val.py)\n        |   `-- crowdpose_test.json\n        `-- images\n            |-- 100000.jpg\n            |-- 100001.jpg\n            |-- 100002.jpg\n            |-- 100003.jpg\n            |-- 100004.jpg\n            |-- 100005.jpg\n            |-- ... \n```\n**For OCHuman data**, please download from [OCHuman download](https://github.com/liruilong940607/OCHumanApi). \nThe ochuman_dir should look like this:\n```\n|-- Click-Pose\n`-- |-- ochuman_dir\n    `-- |-- annotations\n        `-- images\n```\n\n\u003c/details\u003e\n\n\n## 🥳 Run\n\n\n### Train on COCO:\n\n\u003cdetails\u003e\n  \u003csummary\u003eModel-Only\u003c/summary\u003e\n\n```\nexport CLICKPOSE_COCO_PATH=/path/to/your/coco_dir\n python -m torch.distributed.launch --nproc_per_node=4 main.py \\\n    --output_dir \"logs/ClickPose_Model-Only\" \\\n    -c config/clickpose.cfg.py \\\n    --options batch_size=4 epochs=100 lr_drop=80 use_ema=TRUE human_feedback=FLASE feedback_loop_NOC_test=FALSE feedback_inference=FALSE only_correction=FALSE \\\n    --dataset_file=\"coco\" \n```\n\u003c/details\u003e\n\n\n\u003cdetails\u003e\n  \u003csummary\u003eNeural Interactive\u003c/summary\u003e\n\n```\nexport CLICKPOSE_COCO_PATH=/path/to/your/coco_dir\n python -m torch.distributed.launch --nproc_per_node=4 main.py \\\n    --output_dir \"logs/ClickPose_Neural_Interactive\" \\\n    -c config/clickpose.cfg.py \\\n    --options batch_size=4 epochs=100 lr_drop=80 use_ema=TRUE human_feedback=TRUE feedback_loop_NOC_test=FALSE feedback_inference=FALSE only_correction=FALSE \\\n    --dataset_file=\"coco\"\n```\n\u003c/details\u003e\n\n\n\n### Evaluation on COCO:\n\n\u003cdetails\u003e\n  \u003csummary\u003eModel-Only\u003c/summary\u003e\n\n```\nexport CLICKPOSE_COCO_PATH=/path/to/your/coco_dir\n python -m torch.distributed.launch --nproc_per_node=4 main.py \\\n    --output_dir \"logs/ClickPose_Model-Only_eval\" \\\n    -c config/clickpose.cfg.py \\\n    --options batch_size=4 epochs=100 lr_drop=80 use_ema=TRUE human_feedback=FLASE feedback_loop_NOC_test=FALSE feedback_inference=FALSE only_correction=FALSE \\\n    --dataset_file=\"coco\" \\\n    --pretrain_model_path \"./models/ClickPose_model_only_R50.pth\" \\\n    --eval\n```\n\u003c/details\u003e\n\n\n\u003cdetails\u003e\n  \u003csummary\u003eNeural Interactive-NoC metric\u003c/summary\u003e\n\n```\nexport CLICKPOSE_COCO_PATH=/path/to/your/coco_dir\nexport CLICKPOSE_NoC_Test=\"TRUE\"\nexport CLICKPOSE_SAVE_PATH = \"./NoC_95_coco.json\"\nexport NoC_thr = 0.95\n    python -m torch.distributed.launch --nproc_per_node=1 --master_port 3458 main.py \\\n    --output_dir \"logs/ClickPose_Neural_Interactive_eval\" \\\n    -c config/clickpose.cfg.py \\\n    --options batch_size=1 epochs=100 lr_drop=80 use_ema=TRUE human_feedback=TRUE feedback_loop_NOC_test=TRUE feedback_inference=TRUE only_correction=FALSE num_select=20 \\\n    --dataset_file=\"coco\" \\\n    --pretrain_model_path \"./models/ClickPose_interactive_R50.pth\" \\\n    --eval\n```\n\u003c/details\u003e\n\n\n\u003cdetails\u003e\n  \u003csummary\u003eNeural Interactive-AP metric\u003c/summary\u003e\n\n```\nexport CLICKPOSE_COCO_PATH=/path/to/your/coco_dir\nexport CLICKPOSE_NoC_Test=\"TRUE\"\nfor CLICKPOSE_Click_Number in {1..17}\ndo\n    python -m torch.distributed.launch --nproc_per_node=4 --master_port 3458 main.py \\\n    --output_dir \"logs/ClickPose_Neural_Interactive_eval\" \\\n    -c config/clickpose.cfg.py \\\n    --options batch_size=4 epochs=100 lr_drop=80 use_ema=TRUE human_feedback=TRUE feedback_loop_NOC_test=FALSE feedback_inference=TRUE only_correction=FALSE num_select=20 \\\n    --dataset_file=\"coco\" \\\n    --pretrain_model_path \"./models/ClickPose_interactive_R50.pth\" \\\n    --eval\ndone\n\n\n```\n\u003c/details\u003e\n\n\n\n### Evaluation on Human-Art:\n\n\u003cdetails\u003e\n  \u003csummary\u003eModel-Only\u003c/summary\u003e\n\n```\nexport CLICKPOSE_HumanArt_PATH=/path/to/your/humanart_dir\n python -m torch.distributed.launch --nproc_per_node=4 main.py \\\n    --output_dir \"logs/ClickPose_Model-Only_eval\" \\\n    -c config/clickpose.cfg.py \\\n    --options batch_size=4 epochs=100 lr_drop=80 use_ema=TRUE human_feedback=FLASE feedback_loop_NOC_test=FALSE feedback_inference=FALSE only_correction=FALSE \\\n    --dataset_file=\"humanart\" \\\n    --pretrain_model_path \"./models/ClickPose_model_only_R50.pth\" \\\n    --eval\n```\n\u003c/details\u003e\n\n\n\u003cdetails\u003e\n  \u003csummary\u003eNeural Interactive-NoC metric\u003c/summary\u003e\n\n```\nexport CLICKPOSE_HumanArt_PATH=/path/to/your/humanart_dir\nexport CLICKPOSE_NoC_Test=\"TRUE\"\nexport CLICKPOSE_SAVE_PATH = \"./NoC_95_humanart.json\"\nexport NoC_thr = 0.95\n    python -m torch.distributed.launch --nproc_per_node=1 --master_port 3458 main.py \\\n    --output_dir \"logs/ClickPose_Neural_Interactive_eval\" \\\n    -c config/clickpose.cfg.py \\\n    --options batch_size=1 epochs=100 lr_drop=80 use_ema=TRUE human_feedback=TRUE feedback_loop_NOC_test=TRUE feedback_inference=TRUE only_correction=FALSE num_select=20 \\\n    --dataset_file=\"humanart\" \\\n    --pretrain_model_path \"./models/ClickPose_interactive_R50.pth\" \\\n    --eval\n```\n\u003c/details\u003e\n\n\n\u003cdetails\u003e\n  \u003csummary\u003eNeural Interactive-AP metric\u003c/summary\u003e\n\n```\nexport CLICKPOSE_HumanArt_PATH=/path/to/your/humanart_dir\nexport CLICKPOSE_NoC_Test=\"TRUE\"\nfor CLICKPOSE_Click_Number in {1..17}\ndo\n    python -m torch.distributed.launch --nproc_per_node=4 --master_port 3458 main.py \\\n    --output_dir \"logs/ClickPose_Neural_Interactive_eval\" \\\n    -c config/clickpose.cfg.py \\\n    --options batch_size=4 epochs=100 lr_drop=80 use_ema=TRUE human_feedback=TRUE feedback_loop_NOC_test=FALSE feedback_inference=TRUE only_correction=FALSE num_select=20 \\\n    --dataset_file=\"humanart\" \\\n    --pretrain_model_path \"./models/ClickPose_interactive_R50.pth\" \\\n    --eval\ndone\n\n\n```\n\u003c/details\u003e\n\n\n\n### Evaluation on OCHuman:\n\n\u003cdetails\u003e\n  \u003csummary\u003eModel-Only\u003c/summary\u003e\n\n```\nexport CLICKPOSE_OCHuman_PATH=/path/to/your/ochuman_dir\n python -m torch.distributed.launch --nproc_per_node=4 main.py \\\n    --output_dir \"logs/ClickPose_Model-Only_eval\" \\\n    -c config/clickpose.cfg.py \\\n    --options batch_size=4 epochs=100 lr_drop=80 use_ema=TRUE human_feedback=FLASE feedback_loop_NOC_test=FALSE feedback_inference=FALSE only_correction=FALSE \\\n    --dataset_file=\"ochuman\" \\\n    --pretrain_model_path \"./models/ClickPose_model_only_R50.pth\" \\\n    --eval\n```\n\u003c/details\u003e\n\n\n\u003cdetails\u003e\n  \u003csummary\u003eNeural Interactive-NoC metric\u003c/summary\u003e\n\n```\nexport CLICKPOSE_OCHuman_PATH=/path/to/your/ochuman_dir\nexport CLICKPOSE_NoC_Test = \"TRUE\"\nexport CLICKPOSE_SAVE_PATH = \"./NoC_95_ochuman.json\"\nexport NoC_thr = 0.95\n    python -m torch.distributed.launch --nproc_per_node=1 --master_port 3458 main.py \\\n    --output_dir \"logs/ClickPose_Neural_Interactive_eval\" \\\n    -c config/clickpose.cfg.py \\\n    --options batch_size=1 epochs=100 lr_drop=80 use_ema=TRUE human_feedback=TRUE feedback_loop_NOC_test=TRUE feedback_inference=TRUE only_correction=FALSE num_select=20 \\\n    --dataset_file=\"ochuman\" \\\n    --pretrain_model_path \"./models/ClickPose_interactive_R50.pth\" \\\n    --eval\n```\n\u003c/details\u003e\n\n\n\u003cdetails\u003e\n  \u003csummary\u003eNeural Interactive-AP metric\u003c/summary\u003e\n\n```\nexport CLICKPOSE_OCHuman_PATH=/path/to/your/ochuman_dir\nexport CLICKPOSE_NoC_Test=\"TRUE\"\nfor CLICKPOSE_Click_Number in {1..17}\ndo\n    python -m torch.distributed.launch --nproc_per_node=4 --master_port 3458 main.py \\\n    --output_dir \"logs/ClickPose_Neural_Interactive_eval\" \\\n    -c config/clickpose.cfg.py \\\n    --options batch_size=4 epochs=100 lr_drop=80 use_ema=TRUE human_feedback=TRUE feedback_loop_NOC_test=FALSE feedback_inference=TRUE only_correction=FALSE num_select=20 \\\n    --dataset_file=\"ochuman\" \\\n    --pretrain_model_path \"./models/ClickPose_interactive_R50.pth\" \\\n    --eval\ndone\n\n\n```\n\u003c/details\u003e\n\n\n\n\n### Cite Click-Pose\nIf you find this repository useful for your work, please consider citing it as follows:\n\n```\n@inproceedings{yang2023neural,\n  title={Neural Interactive Keypoint Detection},\n  author={Yang, Jie and Zeng, Ailing and Li, Feng and Liu, Shilong and Zhang, Ruimao and Zhang, Lei},\n  booktitle={Proceedings of the IEEE/CVF International Conference on Computer Vision},\n  pages={15122--15132},\n  year={2023}\n}\n```\n\n```\n@inproceedings{yang2022explicit,\n  title={Explicit Box Detection Unifies End-to-End Multi-Person Pose Estimation},\n  author={Yang, Jie and Zeng, Ailing and Liu, Shilong and Li, Feng and Zhang, Ruimao and Zhang, Lei},\n  booktitle={The Eleventh International Conference on Learning Representations},\n  year={2022}\n}\n```\n\n\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fidea-research%2Fclick-pose","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fidea-research%2Fclick-pose","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fidea-research%2Fclick-pose/lists"}