I've been trying to use alexnet to accomplish the task of having multiple labels in an image instead of just one for the highest confidence. i.e. multiple_objects
I've searched through the deep learning tool box and haven't found any examples that cover more than one object while using a neural network. Some examples from the computer vision show tracking a object in motion, but I want the detection to be done regardless if the object is moving, i.e. human, dog, cat, all in the same frame and labeled. The reason I want to do this is because my project requires that all animals + humans be detected with their correct labeling, not person 1 person 2, etc. Here is the example code that I want to go based off of before I spend any time going further.
camera = webcam;
nnet = alexnet;
picture = camera.snapshot;
picture = imresize(picture,[227,227]);
label = classify(nnet, picture);
If I missed a helpful link or video please refer to me to the correct place. Literally almost every example I've seen shows a perfect scenerio where only obe object is being detected, I need all objects in my classification layer to be picked up when seen in the frame as the problem description entails, thank you for your patience and please let me clarify anything if needed.