FaceRecognizer limitations?
Hello,
I would like to use FaceRecognizer to distinguish between specific faces. Accuracy is important to me and I'm making use of eigenfaces as I think it might be the best for this task(?).
So far, in my tests, it seems to be working well, but I would like to know how this will scale up. What if I wanted to distiguish between 1000's of faces? What about 100,000? Is it possible or will I hit memory/speed limits. My main question is:
How much memory does the model typically use up, per face?
Thanks much!
Well i kind of tried what you are asking with about 100 faces,the problem is;loading time of the model and the file size will increase rapidly thus making you wait alot for the model to load into memory.Another thing to take into account aswell is the sample count per face,i don't think you can give it a go with thousands of faces because of the file size of the model would grow beyond reason.
In that case, how would one go about creating a face-recognition system that can recognize 1000's of faces? Are there known techniques?
One possibility that I thought of is to distribute into several models (possibly on different hardware instances), each of which has a certain number of faces... Then, all would need to be run and the one with the best match wins. Does this seem reasonable?
Thanks!