Mars-ViT Pretrained Models

Fang, Jichao

doi:10.5281/zenodo.18180801

Published January 8, 2026 | Version v1

Model Open

Mars-ViT Pretrained Models

Fang, Jichao (Contact person)¹

1. Northern Illinois University

This repository holds model weights described in paper "A Domain-Specific Vision Foundation Model for Mars: Self-Supervised Learning for Planetary-Scale Science Discovery".

Files

README.md

Files (427.6 MB)

Name	Size	Download all
mars-mae-dino-vit-base-v1.pth md5:a2b6c983546329a269f3e512dc2a762e	341.7 MB	Download
mars-mae-dino-vit-small-v1.pth md5:f04cf7e7d44cc89a84af6971aa6d7c84	85.9 MB	Download
README.md md5:f57f7098b85617411e4e3749b7d92295	1.2 kB	Preview Download

Additional details

Submitted: 2025

Repository URL: https://huggingface.co/jfang/mars-vit-base-mae
Programming language: Python
Development Status: Active

Views

Downloads

Show more details

	All versions	This version
Views	54	54
Downloads	4	4
Data volume	769.3 MB	769.3 MB

More info on how stats are collected....

DOI

Resource type

Model

Publisher

Zenodo

License: Creative Commons Attribution 4.0 International

The Creative Commons Attribution license allows re-distribution and re-use of a licensed work on the condition that the creator is appropriately credited. Read more

Technical metadata

Created: January 8, 2026
Modified: January 8, 2026

README.md

Files (427.6 MB)

Dates

Software

Mars-ViT Pretrained Models

Authors/Creators

Description

Files

README.md

Files (427.6 MB)

Additional details

Dates

Software