EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

A Domain‐Specific Vision Foundation Model for Mars: Self‐Supervised Learning for Planetary‐Scale Science Discovery

J. Fang, W. Luo, Q. Huang, L. Zhang, M. S. Phillips, V. D. R. Seethi, I. Giannakis

Journal of Geophysical Research: Machine Learning and Computation · 2026-04 · 3 (2)

DOI: 10.1029/2025jh000827 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Geophysical Research: Machine Learning and Computation, online erschienen am 25.04.2026. Verfasst von J. Fang und 6 weiteren Autor:innen. Die Arbeit nennt 71 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Aeronautics and Space Administration.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Planetary missions have produced a rapidly growing archive of high‐resolution imagery, yet only a small fraction has been examined in detail. Foundation models (FMs) offer a scalable path by learning transferable representations from unlabeled data, but most vision backbones are trained on Internet imagery or Earth observation data sets that differ substantially from planetary surfaces. We present a CTX‐specific Mars vision FM based on Vision Transformer pretrained with self‐supervised learning on millions of Mars Reconnaissance Orbiter (MRO) Context Camera (CTX) images. Across linear evaluation and fine‐tuning for classification, detection, and segmentation, the CTX‐pretrained encoder consistently outperforms strong baseline FMs. The same embeddings enable rapid similarity search across millions of CTX images, reducing the manual effort required to locate geomorphic features and accelerating exploratory mapping at planetary scale.

Förderung

  • National Aeronautics and Space Administration

Zitieren

APA

Fang, J., Luo, W., Huang, Q., Zhang, L., Phillips, M. S., Seethi, V. D. R., & Giannakis, I. (2026). A Domain‐Specific Vision Foundation Model for Mars: Self‐Supervised Learning for Planetary‐Scale Science Discovery. Journal of Geophysical Research: Machine Learning and Computation, 3(2). https://doi.org/10.1029/2025jh000827

Vancouver

Fang J, Luo W, Huang Q, Zhang L, Phillips MS, Seethi VDR, et al. A Domain‐Specific Vision Foundation Model for Mars: Self‐Supervised Learning for Planetary‐Scale Science Discovery. Journal of Geophysical Research: Machine Learning and Computation. 2026;3(2). doi:10.1029/2025jh000827

BibTeX
@article{Fang2026,
  author = {Fang, J. and Luo, W. and Huang, Q. and Zhang, L. and Phillips, M. S. and Seethi, V. D. R. and Giannakis, I.},
  title = {A Domain‐Specific Vision Foundation Model for Mars: Self‐Supervised Learning for Planetary‐Scale Science Discovery},
  journal = {Journal of Geophysical Research: Machine Learning and Computation},
  year = {2026},
  volume = {3},
  number = {2},
  doi = {10.1029/2025jh000827}
}
Kostenlos starten →