Compositional GAN: Learning Conditional Image Composition
arXiv: Computer Vision and Pattern Recognition, Volume abs/1807.07560, 2018.
Generative Adversarial Networks (GANs) can produce images of surprising complexity and realism, but are generally modeled to sample from a single latent source ignoring the explicit spatial interaction between multiple entities that could be present in a scene. Capturing such complex interactions between different objects in the world, in...More
PPT (Upload PPT)