MarsSQE: Stereo Quality Enhancement for Martian Images Using Bi-level Cross-view Attention

TL;DR

MarsSQE is a novel stereo image enhancement method that leverages high cross-view correlations in Martian images using a bi-level attention network, improving quality after lossy compression.

Contribution

Introduces the first dataset of stereo Martian images and a bi-level cross-view attention network for stereo quality enhancement.

Findings

Effective reduction of compression artifacts in Martian images

High cross-view correlations enable better enhancement

First dataset of stereo Martian images created

Abstract

Stereo images captured by Mars rovers are transmitted after lossy compression due to the limited bandwidth between Mars and Earth. Unfortunately, this process results in undesirable compression artifacts. In this paper, we present a novel stereo quality enhancement approach for Martian images, named MarsSQE. First, we establish the first dataset of stereo Martian images. Through extensive analysis of this dataset, we observe that cross-view correlations in Martian images are notably high. Leveraging this insight, we design a bi-level cross-view attention-based quality enhancement network that fully exploits these inherent cross-view correlations. Specifically, our network integrates pixel-level attention for precise matching and patch-level attention for broader contextual information. Experimental results demonstrate the effectiveness of our MarsSQE approach.