Skip to content

Latest commit

 

History

History
17 lines (10 loc) · 1.11 KB

readme.md

File metadata and controls

17 lines (10 loc) · 1.11 KB

Demo

Pretrained weights on UBFC-rPPG dataset and demo code are in the demo folder. The packages in ../requirement.txt should be installed. A additional facenet_pytorch library for face detection should be installed.

You can download my face video here and save the test video in the demo folder. The test face video was recorded by a Logitech C920 webcam and OBS studio with the lossless recording setup.

In the demo folder, simply run

python demo.py

The measured rPPG signals will be shown in results.png as below, and the heart rate will be printed.

Alt text

You can also use your face video to test the demo. You just need to change line 10 in demo.py with your own video path. To improve the running speed, in the face recognition step, we only detect the face in the first frame and keep the bounding box fixed in the following frames. Therefore, please keep stationary in your face video.