Image Super-Resolution via Deterministic-Stochastic Synthesis and Local Statistical Rectification

TL;DR

This paper introduces a novel approach to single image super-resolution by decomposing the high-resolution image into deterministic and stochastic components, then synthesizing and fusing them with deep learning techniques to outperform existing methods.

Contribution

It proposes a new model that separates high-resolution images into deterministic and stochastic parts, using tailored neural networks and local statistical rectification for improved super-resolution.

Findings

Outperforms state-of-the-art super-resolution methods in quantitative metrics.

Demonstrates effectiveness of combining deterministic and stochastic components.

User study confirms visual quality improvements.

Abstract

Single image superresolution has been a popular research topic in the last two decades and has recently received a new wave of interest due to deep neural networks. In this paper, we approach this problem from a different perspective. With respect to a downsampled low resolution image, we model a high resolution image as a combination of two components, a deterministic component and a stochastic component. The deterministic component can be recovered from the low-frequency signals in the downsampled image. The stochastic component, on the other hand, contains the signals that have little correlation with the low resolution image. We adopt two complementary methods for generating these two components. While generative adversarial networks are used for the stochastic component, deterministic component reconstruction is formulated as a regression problem solved using deep neural networks. Since the deterministic component exhibits clearer local orientations, we design novel loss functions tailored for such properties for training the deep regression network. These two methods are first applied to the entire input image to produce two distinct high-resolution images. Afterwards, these two images are fused together using another deep neural network that also performs local statistical rectification, which tries to make the local statistics of the fused image match the same local statistics of the groundtruth image. Quantitative results and a user study indicate that the proposed method outperforms existing state-of-the-art algorithms with a clear margin.