MMLSv2: A Multimodal Dataset for Martian Landslide Detection in Remote Sensing Imagery

Sidike Paheding, Abel Reyes-Angulo, Leo Thomas Ramos, Angel Domingo Sappa, Rajaneesh A., Hiral P. B., Sajin Kumar K. S., Thomas Oommen

Keywords: Martian landslides, Semantic segmentation, Multimodal remote sensing

TL;DR: MMLSv2 is a 7-band multimodal Martian landslide segmentation dataset with a geographically isolated test split that enables rigorous evaluation of spatial generalization and robustness under domain shift.

Abstract: We present MMLSv2, a dataset for landslide segmentation on Martian surfaces. MMLSv2 consists of multimodal imagery with seven bands: RGB, digital elevation model, slope, thermal inertia, and grayscale channels. MMLSv2 comprises 664 images distributed across training, validation, and test splits. In addition, an isolated test set of 276 images from a geographically disjoint region of the base dataset is released to evaluate spatial generalization. Experiments conducted with multiple segmentation models show that the dataset supports stable training and achieves competitive performance, while still posing challenges in fragmented, elongated, and small-scale landslide regions. Evaluation on the isolated test set yields a noticeable performance drop, indicating increased difficulty and underscoring its value for assessing model robustness and generalization beyond standard in-distribution settings. Dataset and code will be released upon acceptance.

Email Sharing: We authorize the sharing of all author emails with Program Chairs.

Data Release: We authorize the release of our submission and author names to the public in the event of acceptance.

Submission Number: 2