Given a pair of images from CIFAR10, average the two images, feed them as input to a network and output the two separate images that were used in the average input.
I am currently using conditional GAN and autoencoders to achieve this task. But have not achieved the desired results till now, the model is being trained but I wanted to know if this is the right direction to solve this task...
This is the output that I am getting till now...
I am afraid if this problem can be solved. What are you getting at the output?