REMAC: Reference-Based Martian Asymmetrical Image Compression

TL;DR

This work introduces REMAC, a reference-based, asymmetrical image compression framework tailored for Martian data with severely constrained encoders. By exploiting strong inter-image and intra-image similarities among Martian images, REMAC shifts computation to the decoder via a reference-guided entropy module and a ref-decoder, while using a deep multi-scale ref-decoder to model long-range dependencies. A latent feature recycling mechanism further reduces encoder workload during inference. Empirical results on Martian imagery show significant encoder-complexity reductions and BD-PSNR gains, highlighting REMAC’s practical potential for efficient Mars-to-Earth communications.

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; 2) They do not utilize the strong \textit{inter-image} similarities across Martian images to advance image compression performance. Motivated by our empirical analysis of the strong \textit{intra-} and \textit{inter-image} 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 \textit{inter-image} 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 \textit{intra-image} similarities, the ref-decoder adopts a deep, multi-scale architecture with enlarged receptive field size to model long-range spatial dependencies. Additionally, 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.

Figures (20)

• Figure 1: The characteristics of Martian and Earth image compression.
• Figure 2: Comparison of representative Martian (left) and Earth (right) images.
• Figure 3: The distributions of PSNR values on Earth and Martian images with Gaussian noise levels of $\sigma=70$, $\sigma=90$, and $\sigma=110$.
• Figure 4: The histograms of Euclidean distances of GLCM features for Martian and Earth images.
• Figure 5: The violin charts of mean values in color channels across different color spaces.