Less is More: Skim Transformer for Light Field Image Super-resolution

TL;DR

This paper introduces the Skim Transformer and SkimLFSR, a lightweight, disparity-aware network for light field image super-resolution that achieves state-of-the-art results with fewer parameters by selectively attending to relevant sub-aperture images.

Contribution

The paper proposes a novel multi-branch Skim Transformer architecture that reduces redundancy and improves efficiency in light field super-resolution by focusing on specific disparity ranges.

Findings

Achieves state-of-the-art super-resolution performance with only 67% of previous parameters.

Demonstrates effective disparity-aware attention behavior guided by predefined SAI subsets.

Shows strong generalizability across different angular resolutions without retraining.

Abstract

A light field image captures scenes through its micro-lens array, providing a rich representation that encompasses spatial and angular information. While this richness comes at significant data redundancy, most existing methods tend to indiscriminately utilize all the information from sub-aperture images (SAIs) in an attempt to harness every visual cue regardless of their disparity significance. However, this paradigm inevitably leads to disparity entanglement, a fundamental cause of inefficiency in light field image processing. To address this limitation, we introduce the Skim Transformer, a novel architecture inspired by the "less is more" philosophy. It features a multi-branch structure where each branch is dedicated to a specific disparity range by constructing its attention score matrix over a skimmed subset of SAIs, rather than all of them. Building upon it, we present SkimLFSR, an efficient yet powerful network for light field image super-resolution. Requiring only 67% of the prior leading method's parameters}, SkimLFSR achieves state-of-the-art results surpassing the best existing method by 0.63 dB and 0.35 dB PSNR at the 2x and 4x tasks, respectively. Through in-depth analyses, we reveal that SkimLFSR, guided by the predefined skimmed SAI sets as prior knowledge, demonstrates distinct disparity-aware behaviors in attending to visual cues. Last but not least, we conduct an experiment to validate SkimLFSR's generalizability across different angular resolutions, where it achieves competitive performance on a larger angular resolution without any retraining or major network modifications. These findings highlight its effectiveness and adaptability as a promising paradigm for light field image processing.