Physics-Based Generative Adversarial Models for Image Restoration and Beyond
This addresses image quality issues in low-level vision tasks for applications like photography and computer vision, but it is incremental as it builds on existing GAN methods.
The authors tackled ill-posed image restoration problems like deblurring and dehazing by proposing a physics model constrained learning algorithm within a GAN framework, which performs favorably against state-of-the-art methods in experiments.
We present an algorithm to directly solve numerous image restoration problems (e.g., image deblurring, image dehazing, image deraining, etc.). These problems are highly ill-posed, and the common assumptions for existing methods are usually based on heuristic image priors. In this paper, we find that these problems can be solved by generative models with adversarial learning. However, the basic formulation of generative adversarial networks (GANs) does not generate realistic images, and some structures of the estimated images are usually not preserved well. Motivated by an interesting observation that the estimated results should be consistent with the observed inputs under the physics models, we propose a physics model constrained learning algorithm so that it can guide the estimation of the specific task in the conventional GAN framework. The proposed algorithm is trained in an end-to-end fashion and can be applied to a variety of image restoration and related low-level vision problems. Extensive experiments demonstrate that our method performs favorably against the state-of-the-art algorithms.