Robust Conditional Generative Adversarial Networks
It addresses robustness for real-world applications of conditional image generation, which is an incremental improvement over existing cGANs.
The paper tackles the problem of conditional GANs being unreliable due to noise sensitivity, introducing RoCGAN to improve robustness, and shows it outperforms state-of-the-art cGANs by a large margin across domains like natural scenes and faces.
Conditional generative adversarial networks (cGAN) have led to large improvements in the task of conditional image generation, which lies at the heart of computer vision. The major focus so far has been on performance improvement, while there has been little effort in making cGAN more robust to noise. The regression (of the generator) might lead to arbitrarily large errors in the output, which makes cGAN unreliable for real-world applications. In this work, we introduce a novel conditional GAN model, called RoCGAN, which leverages structure in the target space of the model to address the issue. Our model augments the generator with an unsupervised pathway, which promotes the outputs of the generator to span the target manifold even in the presence of intense noise. We prove that RoCGAN share similar theoretical properties as GAN and experimentally verify that our model outperforms existing state-of-the-art cGAN architectures by a large margin in a variety of domains including images from natural scenes and faces.