Coupled Generative Adversarial Networks

NeurIPS 2016  ·  Ming-Yu Liu, Oncel Tuzel ·

We propose coupled generative adversarial network (CoGAN) for learning a joint distribution of multi-domain images. In contrast to the existing approaches, which require tuples of corresponding images in different domains in the training set, CoGAN can learn a joint distribution without any tuple of corresponding images. It can learn a joint distribution with just samples drawn from the marginal distributions. This is achieved by enforcing a weight-sharing constraint that limits the network capacity and favors a joint distribution solution over a product of marginal distributions one. We apply CoGAN to several joint distribution learning tasks, including learning a joint distribution of color and depth images, and learning a joint distribution of face images with different attributes. For each task it successfully learns the joint distribution without any tuple of corresponding images. We also demonstrate its applications to domain adaptation and image transformation.

PDF Abstract NeurIPS 2016 PDF NeurIPS 2016 Abstract

Datasets


Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Image-to-Image Translation Cityscapes Labels-to-Photo CoGAN Class IOU 0.06 # 3
Per-class Accuracy 10% # 3
Per-pixel Accuracy 40% # 10
Image-to-Image Translation Cityscapes Photo-to-Labels CoGAN Per-pixel Accuracy 45% # 4
Per-class Accuracy 11% # 4
Class IOU 0.08 # 3

Methods


No methods listed for this paper. Add relevant methods here