{"id":29093321,"url":"https://github.com/nvidia-ai-iot/deepstream_pose_estimation","last_synced_at":"2025-06-28T08:08:08.812Z","repository":{"id":37702916,"uuid":"312384791","full_name":"NVIDIA-AI-IOT/deepstream_pose_estimation","owner":"NVIDIA-AI-IOT","description":"This is a DeepStream application to demonstrate a human pose estimation pipeline.","archived":false,"fork":false,"pushed_at":"2024-06-27T11:08:17.000Z","size":87853,"stargazers_count":289,"open_issues_count":4,"forks_count":66,"subscribers_count":15,"default_branch":"master","last_synced_at":"2024-06-27T16:12:07.200Z","etag":null,"topics":["computer-vision","deepstream","human-pose-estimation","jetson","real-time","tensorrt","tesla"],"latest_commit_sha":null,"homepage":"","language":"C++","has_issues":false,"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":"2020-11-12T20:11:55.000Z","updated_at":"2024-06-27T11:08:22.000Z","dependencies_parsed_at":"2022-09-15T12:02:07.341Z","dependency_job_id":null,"html_url":"https://github.com/NVIDIA-AI-IOT/deepstream_pose_estimation","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/NVIDIA-AI-IOT/deepstream_pose_estimation","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/NVIDIA-AI-IOT%2Fdeepstream_pose_estimation","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/NVIDIA-AI-IOT%2Fdeepstream_pose_estimation/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/NVIDIA-AI-IOT%2Fdeepstream_pose_estimation/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/NVIDIA-AI-IOT%2Fdeepstream_pose_estimation/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/NVIDIA-AI-IOT","download_url":"https://codeload.github.com/NVIDIA-AI-IOT/deepstream_pose_estimation/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/NVIDIA-AI-IOT%2Fdeepstream_pose_estimation/sbom","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":262396520,"owners_count":23304448,"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":["computer-vision","deepstream","human-pose-estimation","jetson","real-time","tensorrt","tesla"],"created_at":"2025-06-28T08:07:51.777Z","updated_at":"2025-06-28T08:08:08.794Z","avatar_url":"https://github.com/NVIDIA-AI-IOT.png","language":"C++","funding_links":[],"categories":[],"sub_categories":[],"readme":"# ------------------------------------------------------\n# This sample application is no longer maintained\n# ------------------------------------------------------\n\n# DeepStream Human Pose Estimation\n\nHuman pose estimation is the computer vision task of estimating the configuration (‘the pose’) of the human body by localizing certain key points on a body within a video or a photo. The following application serves as a reference to deploy custom pose estimation models with DeepStream 5.0 using the [TRTPose](https://github.com/NVIDIA-AI-IOT/trt_pose) project as an example. \n\nA detailed deep-dive NVIDIA Developer blog is available [here](https://developer.nvidia.com/blog/creating-a-human-pose-estimation-application-with-deepstream-sdk/?ncid=so-link-52952-vt24\u0026sfdcid=EM08#cid=em08_so-link_en-us).\n\u003c!--\u003cimg src=\"images/input.gif\" width=\"300\"/\u003e \u003cimg src=\"images/auxillary.png\" width=\"100\"/\u003e \u003cimg src=\"images/output.gif\" width=\"300\"/\u003e--\u003e\n\n\u003ctable\u003e\n  \u003ctr\u003e\n    \u003ctd\u003eInput Video Source\u003c/td\u003e\n     \u003ctd\u003e\u003c/td\u003e\n     \u003ctd\u003eOutput Video\u003c/td\u003e\n  \u003c/tr\u003e\n  \u003ctr\u003e\n    \u003ctd valign=\"top\"\u003e\u003cimg src=\"images/input.gif\"\u003e\u003c/td\u003e\n    \u003ctd valign=\"center\"\u003e\u003cimg src=\"images/auxillary.png\" width=\"100\"\u003e\u003c/td\u003e\n    \u003ctd valign=\"top\"\u003e\u003cimg src=\"images/output.gif\"\u003e\u003c/td\u003e\n  \u003c/tr\u003e\n \u003c/table\u003e\n\n\n## Prerequisites\nYou will need \n1. DeepStreamSDK 5.0\n2. CUDA 10.2\n3. TensorRT 7.x\n\n\n## Getting Started:\nTo get started, please follow these steps.\n1. Install [DeepStream](https://developer.nvidia.com/deepstream-sdk) on your platform, verify it is working by running deepstream-app.\n2. Clone the repository preferably in `$DEEPSTREAM_DIR/sources/apps/sample_apps`.\n2. Download the TRTPose [model](https://github.com/NVIDIA-AI-IOT/trt_pose), convert it to ONNX using this [export utility](https://github.com/NVIDIA-AI-IOT/trt_pose/blob/master/trt_pose/utils/export_for_isaac.py), and set its location in the DeepStream configuration file.\n3. Replace the OSD binaries (x86 or Jetson) in `$DEEPSTREAM_DIR/libs` with the ones provided in this repository under `bin/`. Please note that these are not inter-compatible across platforms.\n4. Compile the program\n ```\n  $ cd deepstream-pose-estimation/\n  $ sudo make\n  $ sudo ./deepstream-pose-estimation-app \u003cfile-uri\u003e \u003coutput-path\u003e\n```\n5. The final output is stored in 'output-path' as `Pose_Estimation.mp4`\n\nNOTE: If you do not already have a .trt engine generated from the ONNX model you provided to DeepStream, an engine will be created on the first run of the application. Depending upon the system you’re using, this may take anywhere from 4 to 10 minutes.\n\nFor any issues or questions, please feel free to make a new post on the [DeepStreamSDK forums](https://forums.developer.nvidia.com/c/accelerated-computing/intelligent-video-analytics/deepstream-sdk/).\n\n## References\nCao, 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\nXiao, 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%2Fdeepstream_pose_estimation","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fnvidia-ai-iot%2Fdeepstream_pose_estimation","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fnvidia-ai-iot%2Fdeepstream_pose_estimation/lists"}