Hi, I will really appreciate it if someone points me to an example that shows how I can create two graphs (lets say features from different layers of a trained network) and then make 2 inferences on the same data without having to load the image twice.
I do not know if it is useful, but my program below loads two kinds of graph.
Pascal VOC + WIDER FACE ---> 20 Class detection and face detection
Two types of detection are performed simultaneously using two kinds of graphs.
It works with normal USB camera.