Robust Conditional Generative Adversarial Networks

Abstract

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.

arXiv commentTo appear in ICLR 2019

References in corpus (10)

• Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
• Conditional Generative Adversarial Nets
• Toward Multimodal Image-to-Image Translation
• Adversarial Examples: Attacks and Defenses for Deep Learning
• Generative Image Inpainting with Contextual Attention
• Semi-supervised Learning with GANs: Manifold Invariance with Improved Inference
• Augmenting Supervised Neural Networks with Unsupervised Objectives for Large-scale Image Classification
• Mathematics of Deep Learning
• One Network to Solve Them All --- Solving Linear Inverse Problems using Deep Projection Models
• Fortified Networks: Improving the Robustness of Deep Networks by Modeling the Manifold of Hidden Representations

Cited by in corpus (8)

• A Review on Generative Adversarial Networks: Algorithms, Theory, and Applications
• StressGAN: A Generative Deep Learning Model for 2D Stress Distribution Prediction
• Reconstructing the Noise Manifold for Image Denoising
• Robust GANs against Dishonest Adversaries
• Learning from Mistakes based on Class Weighting with Application to Neural Architecture Search
• On Leveraging Unlabeled Data for Concurrent Positive-Unlabeled Classification and Robust Generation
• Representation learning with CGAN for casual inference
• Modeling and Forecasting Art Movements with CGANs