Mars-Bench Surface Change Detection, MRO HiRISE Before/After Composites

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-change_cls_hirise)
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-change_cls_hirise

About Mars-Bench Surface Change Detection, MRO HiRISE Before/After Composites

Binary change detection on the Martian surface from MRO HiRISE: 100 x 100 pixel composites stack a before image (blue channel) and an after image (green channel) of the same location, after Kerner et al. (2019), and are labeled change or no_change. Mars-Bench created train, validation and test splits (3,103 / 670 / 670 in the paper) because the source had none, and adds 1- to 20-shot training subsets. Use it alongside mb-change_cls_ctx to compare change detection at HiRISE (sub-meter) and CTX (meter-scale) resolution. Parquet shards under data/ with image and integer label, annotation.csv and mapping.json; load with datasets.load_dataset. About 30 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 Kerner et al. (2019).

How to load this dataset in Python

import pandas as pd
df = pd.read_csv("hf://datasets/Mirali33/mb-change_cls_hirise/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 Surface Change Detection, MRO HiRISE Before/After Composites 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