Learning Texture Transformer Network for Light Field Super-Resolution

TL;DR

This paper introduces a Texture Transformer Network-based method to enhance the spatial resolution of light field images, significantly improving image quality in handheld light field cameras.

Contribution

It presents a novel three-module framework combining all-in-focus image generation, texture transformation, and light field prior refinement for super-resolution.

Findings

Achieves 4-6 dB PSNR gain over bicubic resizing

Effectively enhances light field spatial resolution

Outperforms existing super-resolution methods

Abstract

Hand-held light field cameras suffer from low spatial resolution due to the inherent spatio-angular tradeoff. In this paper, we propose a method to improve the spatial resolution of light field images with the aid of the Texture Transformer Network (TTSR). The proposed method consists of three modules: the first module produces an all-in focus high-resolution perspective image which serves as a reference image for the second module, i.e. TTSR, which in turn produces a high-resolution light field. The last module refines the spatial resolution by imposing a light field prior. The results demonstrate around 4 dB to 6 dB PSNR gain over a bicubically resized light field image