REMAC: Reference-Based Martian Asymmetrical Image Compression

2026-01-26

IEEE Transactions on Geoscience and Remote Sensing

AI Summary

This work addresses the challenge of Mars image compression under severe onboard computational constraints and the underutilization of strong inter-image similarities in existing methods. To this end, we propose a reference-driven asymmetric compression framework that shifts computational burden from the Mars-side encoder to the Earth-side decoder. Our approach introduces three key innovations: reference-guided entropy modeling, a multi-scale decoder architecture with large receptive fields, and a latent feature reuse mechanism, which collectively exploit both intra- and inter-image similarities in texture, color, and semantics. Experimental results demonstrate that, compared to the state-of-the-art, our method reduces encoder complexity by 43.51% while achieving a BD-PSNR gain of 0.2664 dB, significantly alleviating onboard computational load without compromising—and indeed enhancing—compression performance.

Technology Category

• Machine Learning: Learning on the Edge & Model Compression
• Computer Vision: Multi-modal Vision
• Data Mining & Knowledge Management: Data Compression

Application Category

• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

To expedite space exploration on Mars, it is indispensable to develop an efficient Martian image compression method for transmitting images through the constrained Mars-to-Earth communication channel. Although the existing learned compression methods have achieved promising results for natural images from Earth, there remain two critical issues that hinder their effectiveness for Martian image compression: 1) they overlook the highly limited computational resources on Mars; and 2) they do not utilize the strong interimage similarities across Martian images to advance image compression performance. Motivated by our empirical analysis of the strong intraimage and interimage similarities from the perspective of texture, color, and semantics, we propose a reference-based Martian asymmetrical image compression (REMAC) approach, which shifts computational complexity from the encoder to the resource-rich decoder and simultaneously improves compression performance. To leverage interimage similarities, we propose a reference-guided entropy module and a ref-decoder that utilize useful information from reference images, reducing redundant operations at the encoder and achieving superior compression performance. To exploit intraimage similarities, the ref-decoder adopts a deep, multiscale architecture with enlarged receptive field size to model long-range spatial dependencies. In addition, we develop a latent feature recycling mechanism to further alleviate the extreme computational constraints on Mars. Experimental results show that REMAC reduces encoder complexity by 43.51% compared to the state-of-the-art method, while achieving a BD-PSNR gain of 0.2664 dB.

Problem

• Martian image compression
• computational constraints
• inter-image similarity
• asymmetrical compression
• space communication

Innovation

• reference-based compression
• asymmetrical image compression
• inter-image similarity
• computational complexity reduction
• Martian image compression