MOMO: Mars Orbital Model Foundation Model for Mars Orbital Applications

TL;DR

MOMO is a novel multi-sensor foundation model for Mars remote sensing that integrates data from various sensors using a new model merging strategy, achieving superior performance on multiple tasks.

Contribution

Introduces MOMO, the first multi-sensor foundation model for Mars, with a novel EVL strategy for effective model merging across sensors.

Findings

MOMO outperforms existing models on Mars-Bench tasks.

EVL strategy improves model stability and generalization.

Model merging at compatible checkpoints enhances multi-resolution data integration.

Abstract

We introduce MOMO, the first multi-sensor foundation model for Mars remote sensing. MOMO uses model merge to integrate representations learned independently from three key Martian sensors (HiRISE, CTX, and THEMIS), spanning resolutions from 0.25 m/pixel to 100 m/pixel. Central to our method is our novel Equal Validation Loss (EVL) strategy, which aligns checkpoints across sensors based on validation loss similarity before fusion via task arithmetic. This ensures models are merged at compatible convergence stages, leading to improved stability and generalization. We train MOMO on a large-scale, high-quality corpus of $\sim 12$ million samples curated from Mars orbital data and evaluate it on 9 downstream tasks from Mars-Bench. MOMO achieves better overall performance compared to ImageNet pre-trained, earth observation foundation model, sensor-specific pre-training, and fully-supervised baselines. Particularly on segmentation tasks, MOMO shows consistent and significant performance improvement. Our results demonstrate that model merging through an optimal checkpoint selection strategy provides an effective approach for building foundation models for multi-resolution data. The model weights, pretraining code, pretraining data, and evaluation code are available at: https://github.com/kerner-lab/MOMO.