AlexeyAB / darknet

YOLOv4 / Scaled-YOLOv4 / YOLO - Neural Networks for Object Detection (Windows and Linux version of Darknet )
http://pjreddie.com/darknet/
Other
21.65k stars 7.96k forks source link

mtcnn on darknet #3923

Open tomriddle54 opened 5 years ago

tomriddle54 commented 5 years ago

Hello again @AlexeyAB I am planning to integrate mtcnn(Multi-task Cascaded Convolutional Network) in darknet. The MTCNN model consists of 3 separate networks: the P-Net, the R-Net, and the O-Net: So my question is can i create a cfg file for this and train my model on it. I think it is possible in darknet image

AlexeyAB commented 5 years ago

Why is 3 different P / R / O networks required?

You can try to combine these 3 networks into 1 cfg-file. But there is no layer for Facial landmark localization, so as I understand you should implement it by yourself.

tomriddle54 commented 5 years ago

I don't know c well but python I know so how can I create layer can you help lets create a face detection on darknet I feel that we can get about 40 fps face detection on 4gb gpu