I am doing an internship project regarding deep learning, and it is a totally new topic for me as I have never studied machine learning in the bachelor's degree courses. I have to implement a GAN that generates a middle image starting from two images, the previous and the next. In particular, the network must work with TACs, therefore it must generate the middle TAC starting from two successive ones. I am studying the basic theory concerning deep learning, neural networks and in particular GANs, but I have not yet fully understood where the network takes in the two starting images. Does the generating network still have to take in only one noise vector or must it also know the two images? Can you recommend any architecture that I can use to get ideas? Do you have any advice?
Thanks in advance