Face recognition with Deepstream

This is a face detection and recognition demo pipeline build on Deepstream.

Requirements

• Deepstream 6.0+
• GStreamer 1.14.5+
• Cuda 11.4+
• NVIDIA driver 470.63.01+
• TensorRT 8+
• Python 3.6+
• Opencv

Follow deepstream official doc to install dependencies.

Deepstream docker is more recommended.

Models

• yolov8-face
• retinaface
• arcface

Alignment

gst-nvinfer-custom

Usage

1 - prepare data

• put the feature file(.npy format) to data/known_faces
• or put face images to data/unknown_faces and run python3 utils/gen_feature.py

  2 - compile

• gst-nvinfer-custom : follow README
• nvdsinfer_customparser for detector post-process

  3 - run

  python3 main.py

References

• NVIDIA-AI-IOT/deepstream_python_apps
• biubug6/Pytorch_Retinaface
• yolov8-face
• arcface