LunarFM: A Shared Multimodal Representation of the Moon's Surface
Marc Girona-Mata, Jakob Gawlikowski, Sumit Goski, Gautier Bardi de Fourtou, Valentin T. Bickel, Ben Moseley, Abigail Calzada-Diaz, Sylvester Kaczmarek, Ra\'ul Ramos-Poll\'an
Why It Matters
What makes this one worth your time
LunarFM provides a comprehensive tool for lunar exploration, enabling more efficient analysis and utilization of lunar resources, which is crucial for future lunar missions and potential colonization.
LunarFM offers a unified multimodal representation of the Moon's surface for diverse scientific and resource mapping tasks.
Summary
The paper introduces LunarFM, a multimodal foundation model that creates a shared representation of the lunar surface by integrating data from six instruments across three lunar missions. This model supports various applications such as similarity search, resource mapping, and geological classification, and is accompanied by a machine-learning-ready dataset and pretrained model.
Key contributions
- Development of a multimodal foundation model for the lunar surface.
- Creation of a machine-learning-ready dataset of co-registered multimodal observations.
- Provision of a pretrained multimodal masked autoencoder and a companion embedding dataset.
Notable insights
- The use of a shared embedding space for diverse lunar data facilitates a wide range of applications from a single model.
- The integration of data from multiple instruments and missions into a unified model is a novel approach in lunar surface analysis.
Possible limitations
- Not stated in the abstract
Abstract
arXiv:2607.22408v1 Announce Type: new Abstract: The renewed global focus on lunar exploration, driven by the prospect of in-situ resource utilization and a sustained human presence on the Moon, has created growing demand for accurate, large-scale characterization of the lunar surface. Although vast quantities of orbital remote-sensing data have been collected, scientific analysis and resource mapping remain fragmented by heterogeneous multiinstrument observations, sparse labels, and bespoke task-specific modelling workflows. Here we introduce LunarFM, a multimodal foundation model that learns a general representation of the lunar surface from diverse orbital measurements. LunarFM assimilates observations from six instruments across three lunar missions, mapping 18 input channels to a shared embedding space. We demonstrate that this embedding space supports a diverse range of downstream applications, including similarity search, few-shot resource mapping, mineral abundance regression, and geological unit classification, enabling efficient scientific investigation and resource-oriented analysis. We provide a machine-learning-ready dataset of co-registered multimodal observations spanning latitudes from 70{\deg}S to 70{\deg}N, a pretrained multimodal masked autoencoder, and a companion embedding dataset providing a joint 768-dimensional representation of lunar surface properties. All code and data are available at https://lunarfm.trillium.tech/