{"id":13644353,"url":"https://github.com/zhouyuchong/face-recognition-deepstream","last_synced_at":"2025-04-21T07:31:01.126Z","repository":{"id":40353428,"uuid":"463082822","full_name":"zhouyuchong/face-recognition-deepstream","owner":"zhouyuchong","description":"Deepstream app use retinaface and arcface for face recognition.","archived":false,"fork":false,"pushed_at":"2022-10-24T08:56:38.000Z","size":582,"stargazers_count":44,"open_issues_count":5,"forks_count":10,"subscribers_count":0,"default_branch":"main","last_synced_at":"2024-02-12T23:13:18.601Z","etag":null,"topics":["arcface","deepstream","face-recognition","retinaface"],"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/zhouyuchong.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}},"created_at":"2022-02-24T09:21:02.000Z","updated_at":"2024-02-06T02:35:12.000Z","dependencies_parsed_at":"2022-07-07T23:09:33.641Z","dependency_job_id":null,"html_url":"https://github.com/zhouyuchong/face-recognition-deepstream","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/zhouyuchong%2Fface-recognition-deepstream","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/zhouyuchong%2Fface-recognition-deepstream/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/zhouyuchong%2Fface-recognition-deepstream/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/zhouyuchong%2Fface-recognition-deepstream/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/zhouyuchong","download_url":"https://codeload.github.com/zhouyuchong/face-recognition-deepstream/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":250014603,"owners_count":21360980,"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":["arcface","deepstream","face-recognition","retinaface"],"created_at":"2024-08-02T01:02:01.770Z","updated_at":"2025-04-21T07:30:56.098Z","avatar_url":"https://github.com/zhouyuchong.png","language":"Python","funding_links":[],"categories":["Face Recognition","Object Detection Applications"],"sub_categories":[],"readme":"# Face recognition with Deepstream 6.1\nThis is a face detection and recognition app build on Deepstream.\n\n## Requirements\n+ Deepstream 6.0+\n+ GStreamer 1.14.5+\n+ Cuda 11.4+\n+ NVIDIA driver 470.63.01+\n+ TensorRT 8+\n+ Python 3.6+\n\nFollow [deepstream](https://docs.nvidia.com/metropolis/deepstream/dev-guide/text/DS_Quickstart.html#dgpu-setup-for-ubuntu) official doc to install dependencies.\n\nDeepstream docker is more recommended.\n\n## Pretrained\nPlease refer to links below for pretrained models and serialized TensorRT engine. Or download from [Google driver](https://drive.google.com/drive/folders/1HTdIhGrKP7JnKY6n8F95mI7SBnx7-4R3).\n+ ~~[yolov5](https://github.com/wang-xinyu/tensorrtx/tree/master/yolov5)~~\n+ [retinaface](https://github.com/wang-xinyu/tensorrtx/tree/master/retinaface)\n+ [arcface](https://github.com/wang-xinyu/tensorrtx/tree/master/arcface)\n\n## Alignment\nthere should be a face alignment before arcface. Use a custom-gst-nvinfer to preprocess the tensor-meta of retinaface. \n\n[Demo custom-gst-nvinfer](https://github.com/zhouyuchong/gst-nvinfer-custom)\n\n## Usage\n### 1 - compile retinaface parse function\nset a larger face detection bbox so that we can do alignment better later.\n```\ncd models/retinaface\nmake\n```\n### 2 - modify all path in config and codes\n+ all path in src/config/face/\n+ line 24 in src/kbds/app/face.py\n+ video path in test script\n\n### 3 - Start kafka service before everything. \nThen set \"ip;port;topic\" in codes.\ne.g.\n```\nDSFace(\"localhost;9092;deepstream\")\n```\nyou can recieve result messages on **deepstream** topic and error messages on **error** topic in same ip\u0026port.\n### 4 - start app\n```\ncd test\npython3 face_test_demo.py\n```\nNotice that only support RTSP stream and Gstreamer format.\n\n### 5 - check result\nimages before and after alignment will be stored in images.\n\n\nmodify codes in test/cal_sim.py. replace with your own face-feature npy file.\n```\npython3 test/cal_sim.py\n```\nit will outputs all score.\n\n\n## References\n+ [wang-xinyu/tensorrtx](https://github.com/wang-xinyu/tensorrtx)\n+ [NVIDIA-AI-IOT/deepstream_python_apps](https://github.com/NVIDIA-AI-IOT/deepstream_python_apps)\n\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fzhouyuchong%2Fface-recognition-deepstream","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fzhouyuchong%2Fface-recognition-deepstream","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fzhouyuchong%2Fface-recognition-deepstream/lists"}