MimicGAN: Corruption-Mimicking for Blind Image Recovery & Adversarial Defense

11/20/2018
by   Rushil Anirudh, et al.
10

Solving inverse problems continues to be a central challenge in computer vision. Existing techniques either explicitly construct an inverse mapping using prior knowledge about the corruption, or learn the inverse directly using a large collection of examples. However, in practice, the nature of corruption may be unknown, and thus it is challenging to regularize the problem of inferring a plausible solution. On the other hand, collecting task-specific training data is tedious for known corruptions and impossible for unknown ones. We present MimicGAN, an unsupervised technique to solve general inverse problems based on image priors in the form of generative adversarial networks (GANs). Using a GAN prior, we show that one can reliably recover solutions to underdetermined inverse problems through a surrogate network that learns to mimic the corruption at test time. Our system successively estimates the corruption and the clean image without the need for supervisory training, while outperforming existing baselines in blind image recovery. We also demonstrate that MimicGAN improves upon recent GAN-based defenses against adversarial attacks and represents one of the strongest test-time defenses available today.

READ FULL TEXT

page 1

page 4

page 7

page 8

page 14

page 15

page 16

research
05/18/2018

An Unsupervised Approach to Solving Inverse Problems using Generative Adversarial Networks

Solving inverse problems continues to be a challenge in a wide array of ...
research
01/23/2020

Reducing the Representation Error of GAN Image Priors Using the Deep Decoder

Generative models, such as GANs, learn an explicit low-dimensional repre...
research
12/16/2019

MimicGAN: Robust Projection onto Image Manifolds with Corruption Mimicking

In the past few years, generative models like Generative Adversarial Net...
research
10/08/2018

Algorithmic Aspects of Inverse Problems Using Generative Models

The traditional approach of hand-crafting priors (such as sparsity) for ...
research
10/07/2021

StyleGAN-induced data-driven regularization for inverse problems

Recent advances in generative adversarial networks (GANs) have opened up...
research
06/08/2023

Solution of physics-based inverse problems using conditional generative adversarial networks with full gradient penalty

The solution of probabilistic inverse problems for which the correspondi...
research
09/17/2020

Integration of AI and mechanistic modeling in generative adversarial networks for stochastic inverse problems

The problem of finding distributions of input parameters for determinist...

Please sign up or login with your details

Forgot password? Click here to reset