Multi-scale interaction network for stereo image super-resolution

TL;DR

This paper introduces a multi-scale interaction network that enhances stereo image super-resolution by exploiting intra-view and cross-view information through novel attention modules and optimal transport matching.

Contribution

The paper proposes a new multi-scale interaction network with specialized attention modules and an optimal transport-based matching method for improved stereo image super-resolution.

Findings

Achieves competitive results outperforming most SOTA methods.

Effective intra-view feature extraction via multi-scale spatial-channel attention.

Accurate cross-view matching using epipolar line-based optimal transport.

Abstract

Stereo image super-resolution aims to generate high-resolution images by leveraging complementary information from binocular systems. Although previous studies have achieved impressive results, the potential of intra-view and cross-view information has not been fully exploited. To address this issue, we propose a novel multi-scale interaction network for stereo image super-resolution. Specifically, we design a Multi-scale Spatial-Channel Attention Module that utilizes multi-scale large separable kernel attention and simple channel attention to improve intra-view feature extraction. Additionally, we propose a Dual-View Epipolar Attention Module, utilizing an optimal transport algorithm to achieve more accurate matching along the epipolar line. Extensive experimental and ablation studies show that our method achieves competitive results that outperform most SOTA methods.