{"id":29093248,"url":"https://github.com/nvidia-ai-iot/trt_pose","last_synced_at":"2025-06-28T08:07:54.345Z","repository":{"id":41192362,"uuid":"199516660","full_name":"NVIDIA-AI-IOT/trt_pose","owner":"NVIDIA-AI-IOT","description":"Real-time pose estimation accelerated with NVIDIA TensorRT","archived":false,"fork":false,"pushed_at":"2022-08-12T00:45:06.000Z","size":2065,"stargazers_count":966,"open_issues_count":136,"forks_count":288,"subscribers_count":41,"default_branch":"master","last_synced_at":"2024-07-31T19:23:21.575Z","etag":null,"topics":["human-pose","human-pose-estimation","jetson","jetson-nano","jetson-xavier","live-demo","pretrained-models","pytorch","real-time","tensorrt","torch2trt"],"latest_commit_sha":null,"homepage":"","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/NVIDIA-AI-IOT.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE.md","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null}},"created_at":"2019-07-29T19:42:14.000Z","updated_at":"2024-07-30T16:16:55.000Z","dependencies_parsed_at":"2022-08-10T01:43:05.812Z","dependency_job_id":null,"html_url":"https://github.com/NVIDIA-AI-IOT/trt_pose","commit_stats":null,"previous_names":[],"tags_count":1,"template":false,"template_full_name":null,"purl":"pkg:github/NVIDIA-AI-IOT/trt_pose","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/NVIDIA-AI-IOT%2Ftrt_pose","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/NVIDIA-AI-IOT%2Ftrt_pose/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/NVIDIA-AI-IOT%2Ftrt_pose/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/NVIDIA-AI-IOT%2Ftrt_pose/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/NVIDIA-AI-IOT","download_url":"https://codeload.github.com/NVIDIA-AI-IOT/trt_pose/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/NVIDIA-AI-IOT%2Ftrt_pose/sbom","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":262396520,"owners_count":23304447,"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":["human-pose","human-pose-estimation","jetson","jetson-nano","jetson-xavier","live-demo","pretrained-models","pytorch","real-time","tensorrt","torch2trt"],"created_at":"2025-06-28T08:07:42.047Z","updated_at":"2025-06-28T08:07:54.307Z","avatar_url":"https://github.com/NVIDIA-AI-IOT.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# trt_pose\n\n\u003e Want to detect hand poses?  Check out the new [trt_pose_hand](http://github.com/NVIDIA-AI-IOT/trt_pose_hand) project for real-time hand pose and gesture recognition!\n\n\u003cimg src=\"https://user-images.githubusercontent.com/4212806/67125332-71a64580-f1a9-11e9-8ee1-e759a38de215.gif\" height=256/\u003e\n\ntrt_pose is aimed at enabling real-time pose estimation on NVIDIA Jetson.  You may find it useful for other NVIDIA platforms as well.  Currently the project includes\n\n- Pre-trained models for human pose estimation capable of running in real time on Jetson Nano.  This makes it easy to detect features like ``left_eye``, ``left_elbow``, ``right_ankle``, etc.\n\n- Training scripts to train on any keypoint task data in [MSCOCO](https://cocodataset.org/#home) format.  This means you can experiment with training trt_pose for keypoint detection tasks other than human pose.\n\nTo get started, follow the instructions below.  If you run into any issues please [let us know](../../issues).\n\n## Getting Started\n\nTo get started with trt_pose, follow these steps.\n\n### Step 1 - Install Dependencies\n\n1. Install PyTorch and Torchvision.  To do this on NVIDIA Jetson, we recommend following [this guide](https://forums.developer.nvidia.com/t/72048)\n\n2. Install [torch2trt](https://github.com/NVIDIA-AI-IOT/torch2trt)\n\n    ```python\n    git clone https://github.com/NVIDIA-AI-IOT/torch2trt\n    cd torch2trt\n    sudo python3 setup.py install --plugins\n    ```\n\n3. Install other miscellaneous packages\n\n    ```python\n    sudo pip3 install tqdm cython pycocotools\n    sudo apt-get install python3-matplotlib\n    ```\n    \n### Step 2 - Install trt_pose\n\n```python\ngit clone https://github.com/NVIDIA-AI-IOT/trt_pose\ncd trt_pose\nsudo python3 setup.py install\n```\n\n### Step 3 - Run the example notebook\n\nWe provide a couple of human pose estimation models pre-trained on the MSCOCO dataset.  The throughput in FPS is shown for each platform\n\n| Model | Jetson Nano | Jetson Xavier | Weights |\n|-------|-------------|---------------|---------|\n| resnet18_baseline_att_224x224_A | 22 | 251 | [download (81MB)](https://drive.google.com/open?id=1XYDdCUdiF2xxx4rznmLb62SdOUZuoNbd) |\n| densenet121_baseline_att_256x256_B | 12 | 101 | [download (84MB)](https://drive.google.com/open?id=13FkJkx7evQ1WwP54UmdiDXWyFMY1OxDU) |\n\nTo run the live Jupyter Notebook demo on real-time camera input, follow these steps\n \n1. Download the model weights using the link in the above table.  \n\n2. Place the downloaded weights in the [tasks/human_pose](tasks/human_pose) directory\n\n3. Open and follow the [live_demo.ipynb](tasks/human_pose/live_demo.ipynb) notebook\n\n    \u003e You may need to modify the notebook, depending on which model you use\n\n## See also\n\n- [trt_pose_hand](http://github.com/NVIDIA-AI-IOT/trt_pose_hand) - Real-time hand pose estimation based on trt_pose\n- [torch2trt](http://github.com/NVIDIA-AI-IOT/torch2trt) - An easy to use PyTorch to TensorRT converter\n\n- [JetBot](http://github.com/NVIDIA-AI-IOT/jetbot) - An educational AI robot based on NVIDIA Jetson Nano\n- [JetRacer](http://github.com/NVIDIA-AI-IOT/jetracer) - An educational AI racecar using NVIDIA Jetson Nano\n- [JetCam](http://github.com/NVIDIA-AI-IOT/jetcam) - An easy to use Python camera interface for NVIDIA Jetson\n\n## References\n\nThe trt_pose model architectures listed above are inspired by the following works, but are not a direct replica.  Please review the open-source code and configuration files in this repository for architecture details.  If you have any questions feel free to reach out.\n\n*  _Cao, Zhe, et al. \"Realtime multi-person 2d pose estimation using part affinity fields.\" Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2017._\n\n*  _Xiao, Bin, Haiping Wu, and Yichen Wei. \"Simple baselines for human pose estimation and tracking.\" Proceedings of the European Conference on Computer Vision (ECCV). 2018._\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fnvidia-ai-iot%2Ftrt_pose","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fnvidia-ai-iot%2Ftrt_pose","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fnvidia-ai-iot%2Ftrt_pose/lists"}