Mars-Bench Curiosity Rover Surface Image Classification, Mastcam and MAHLI (36 classes)

Provider
Mirali Purohit, Hannah Kerner et al. (Mars-Bench, Arizona State University)
License
CC-BY-4.0 (source)
Size
20 files
Format
CSV, Parquet (in sampled files)
Stored at
Hugging Face (https://huggingface.co/datasets/Mirali33/mb-surface_cls)
Access
Free and public. No account with the data provider is needed.
DOI
10.48550/arXiv.2510.24010
Category
Astronomy & Space
Source
https://huggingface.co/datasets/Mirali33/mb-surface_cls

About this dataset

Classification of images taken by the Curiosity rover (Mars Science Laboratory) with its Mastcam and MAHLI cameras, 256 x 256 pixels, revised from the MSL surface datasets of Wagstaff et al. (2018, 2021). The 36 classes mix geology and rover hardware: sand, ground, float rock, layered rock, light-toned veins, close-up rock, drill holes, dust removal tool spots, distant landscape, horizon, night sky, sun, wheel tracks, and instrument views such as APXS, ChemCam and Mastcam calibration targets, CheMin inlet, drill, scoop, turret, wheels and REMS-UV. Splits: 6,580 training, 1,293 validation, 1,594 test, plus few-shot subsets. Used to search and triage the rover's downlinked image archive. Parquet shards under data/, mapping.json for class names; load with datasets. About 175 MB. Part of Mars-Bench (Purohit et al., NeurIPS 2025 Datasets and Benchmarks track, arXiv:2510.24010), 20 standardized Mars datasets released by the authors under CC BY 4.0; cite Mars-Bench and Wagstaff et al. (2018, 2021).

How to load this dataset in Python

import pandas as pd
df = pd.read_csv("hf://datasets/Mirali33/mb-surface_cls/annotation.csv")
df.head()

Data structure

File types in sampled files: CSV, Parquet

Files

Example file: annotation.csv

Analyze this dataset without downloading it

On Scigantic, Mars-Bench Curiosity Rover Surface Image Classification, Mastcam and MAHLI (36 classes) opens in a hosted JupyterLab notebook with the data already mounted, so you can read files where they are stored instead of copying them first. Each notebook includes an AI assistant that can see the mounted files and write and run analysis code. Open the dataset page to launch a notebook, or browse the full catalog.

This page is also available as Markdown and as JSON.

Related Astronomy & Space datasets