I am currently working on undergraduate research to determine hotspots for hand-surface contact. Ideally, I would like to give the model a depth image as input:
Example of synthetic depth image
and return an image mask indicating where the surface was touched:
Example of synthetic contact mask
I have worked with Machine Learning before but am struggling to determine what model I should use. My understanding is that CNNs are typically intended for classification tasks. And while GANs are used to generate new images, they can produce these images independently of an input. Assuming I have a large dataset of depth images and the respective black and white contact mask, what model can be used to efficiently predict a contact mask given an unseen depth image?
