Mars-Bench Crater Binary Segmentation, Mars Odyssey THEMIS (Robbins and Hynek craters)
- Provider
- Mirali Purohit, Hannah Kerner et al. (Mars-Bench, Arizona State University)
- License
- CC-BY-4.0 (source)
- Size
- 15 files
- Format
- Parquet (in sampled files)
- Stored at
- Hugging Face (https://huggingface.co/datasets/Mirali33/mb-crater_binary_seg)
- 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-crater_binary_seg
About this dataset
Binary segmentation of impact craters in 512 x 512 pixel tiles from the Mars Odyssey THEMIS daytime infrared global mosaic, with masks prepared by the Mars-Bench authors from the Robbins and Hynek (2012) global Mars crater database. Pixels are background or crater. Splits are 3,600 training, 900 validation and 900 test tiles, plus 1 to 50 percent training fractions; metadata.zip gives tile locations. Crater mapping underlies surface age estimation from crater counts, so this is a common pretext task for planetary segmentation models. A five-class version by crater morphology is mb-crater_multi_seg. Parquet shards under data/ with image, mask, width, height and labels; load with datasets.load_dataset("Mirali33/mb-crater_binary_seg"). About 1.9 GB. 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 Robbins and Hynek (2012).
How to load this dataset in Python
from datasets import load_dataset
ds = load_dataset("Mirali33/mb-crater_binary_seg")
print(ds)Data structure
File types in sampled files: Parquet
Files
Example file: data/partition_train_0.01x_partition-00000-of-00001.parquet
Analyze this dataset without downloading it
On Scigantic, Mars-Bench Crater Binary Segmentation, Mars Odyssey THEMIS (Robbins and Hynek craters) 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
- ESA/Hubble Full-Resolution Space Images with Captions (2,706 Images)
- Magnetospheric Multiscale (MMS) Level 2 Particle and Field Data, All Four Spacecraft (CDAWeb on HelioCloud)
- Zwicky Transient Facility (ZTF) DR24 Lightcurves, HATS Collection
- Voyager 1 and 2 ISS Images of Europa, Photogrammetrically Controlled (USGS Astrogeology)
- AWE Atmospheric Waves Experiment on the ISS: Mesospheric Temperature and OH Airglow Swaths, Level 0 and 3 (CDAWeb on HelioCloud)
- Open Supernova Catalog: 72,145 Supernovae and Candidates with Types, Redshifts and Hosts
- Mars-Bench HiRISE Landmark Classification (8 classes)
- SESAME Satellite Experiments Simultaneous with Antarctic Measurements: Halley Ionosonde, Fabry-Perot, Riometer, Magnetometer and VLF Key Parameters, 1992-2001 (CDAWeb on HelioCloud)