Keywords: Generative Adversarial Network, GAN, cGAN, Structured Prediction, Fusion.
TL;DR: We propose the fusion discriminator, a novel architecture for incorporating conditional information into the discriminator of GANs for structured prediction tasks.
Abstract: We propose the fusion discriminator, a single unified framework for incorporating conditional information into a generative adversarial network (GAN) for a variety of distinct structured prediction tasks, including image synthesis, semantic segmentation, and depth estimation. Much like commonly used convolutional neural network - conditional Markov random field (CNN-CRF) models, the proposed method is able to enforce higher-order consistency in the model, but without being limited to a very specific class of potentials. The method is conceptually simple and flexible, and our experimental results demonstrate improvement on several diverse structured prediction tasks.