DualSwinFusionSeg: Multimodal Martian Landslide Segmentation via Dual Swin Transformer with Multi-Scale Fusion and UNet++
Computer Vision and Pattern RecognitionMachine Learning
Summary
The authors created a new computer model called DualSwinFusionSeg to find landslides on Mars using images and other types of data like elevation and temperature. Their model uses two separate but connected parts to look at regular pictures and special scientific data, then combines what it learns to better identify landslides. They tested it on a Mars landslide dataset and found it works well even when there isn’t a lot of training data. This helps improve automated mapping for future Mars exploration.
Authors
Shahriar Kabir, Abdullah Muhammed Amimul Ehsan, Istiak Ahmmed Rifti, Md Kaykobad Reza
Abstract
Automated segmentation of Martian landslides, particularly in tectonically active regions such as Valles Marineris,is important for planetary geology, hazard assessment, and future robotic exploration. However, detecting landslides from planetary imagery is challenging due to the heterogeneous nature of available sensing modalities and the limited number of labeled samples. Each observation combines RGB imagery with geophysical measurements such as digital elevation models, slope maps, thermal inertia, and contextual grayscale imagery, which differ significantly in resolution and statistical properties. To address these challenges, we propose DualSwinFusionSeg, a multimodal segmentation architecture that separates modality-specific feature extraction and performs multi-scale cross-modal fusion. The model employs two parallel Swin Transformer V2 encoders to independently process RGB and auxiliary geophysical inputs, producing hierarchical feature representations. Corresponding features from the two streams are fused at multiple scales and decoded using a UNet++ decoder with dense nested skip connections to preserve fine boundary details. Extensive ablation studies evaluate modality contributions, loss functions, decoder architectures, and fusion strategies. Experiments on the MMLSv2 dataset from the PBVS 2026 Mars-LS Challenge show that modality-specific encoders and simple concatenation-based fusion improve segmentation accuracy under limited training data. The final model achieves 0.867 mIoU and 0.905 F1 on the development benchmark and 0.783 mIoU on the held-out test set, demonstrating strong performance for multimodal planetary surface segmentation.