This is my first topic and I hope you can help me out. I have to create a face detection if you wear e.g. head mounted displays. So OpenCV should recognize the mouth and then predict the face, the next step would be to get 3d vertices to recreate the whole face. But for now I would be happy to find a way to recognize the mouth and with that the whole face. What would be the best option?
I am looking at the moment for a way with tensorflow, to create a cnn only with pictures of mouth. Then I would compare it with a pre run face detection from openCV to have a ground truth. But how would I compare these to results then?
Can someone help me out? I’ve been racking my brains all day on this.
Thanks in advance!