This video shows 2 face detectors:
1) OpenCVs Viola Jones implementation (red)
2) YOLOv2 trained on the WIDER FACE dataset (green)
I'm using OpenCV's dnn module for YOLO inference.
It's running on the CPU hence the lagginess.
Check the face detection model zoo with 4 pre trained models:
https://github.com/the-house-of-black...
The github repo with running instructions (using docker), training details and final models for download can be found at:
https://github.com/the-house-of-black...
Watch video Deep learning Face detection: YOLO v2 (darknet) vs Viola Jones online without registration, duration hours minute second in high quality. This video was added by user Fábio Uechi 10 January 2018, don't forget to share it with your friends and acquaintances, it has been viewed on our site 2,00 once and liked it 1 people.