# DualSwinFusionSeg: Multimodal Martian Landslide Segmentation via Dual Swin Transformer with Multi-Scale Fusion and UNet++

arXiv: https://arxiv.org/abs/2603.14132 · HTML: https://mlsanity.com/paper/2603.14132/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 17 of its 27 references that are in the map.
- First posted on arXiv 14 Mar 2026.

## Builds on (most cited first, 17 of 17)

- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [Swin Transformer: Hierarchical Vision Transformer using Shifted Windows](https://mlsanity.com/paper/2103.14030/) (2021)
- [Feature Pyramid Networks for Object Detection](https://mlsanity.com/paper/1612.03144/) (2016)
- [Fully Convolutional Networks for Semantic Segmentation](https://mlsanity.com/paper/1411.4038/) (2014)
- [Squeeze-and-Excitation Networks](https://mlsanity.com/paper/1709.01507/) (2017)
- [Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation](https://mlsanity.com/paper/1802.02611/) (2018)
- [SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformers](https://mlsanity.com/paper/2105.15203/) (2021)
- [CBAM: Convolutional Block Attention Module](https://mlsanity.com/paper/1807.06521/) (2018)
- [UNet++: A Nested U-Net Architecture for Medical Image Segmentation](https://mlsanity.com/paper/1807.10165/) (2018)
- [Unified Perceptual Parsing for Scene Understanding](https://mlsanity.com/paper/1807.10221/) (2018)
- [Swin Transformer V2: Scaling Up Capacity and Resolution](https://mlsanity.com/paper/2111.09883/) (2021)
- [Albumentations: fast and flexible image augmentations](https://mlsanity.com/paper/1809.06839/) (2018)
- [Tversky loss function for image segmentation using 3D fully convolutional deep networks](https://mlsanity.com/paper/1706.05721/) (2017)
- [Landslide4Sense: Reference Benchmark Data and Deep Learning Models for Landslide Detection](https://mlsanity.com/paper/2206.00515/) (2022)
- [MMSFormer: Multimodal Transformer for Material and Semantic Segmentation](https://mlsanity.com/paper/2309.04001/) (2023)
- [TransLandSeg: A Transfer Learning Approach for Landslide Semantic Segmentation Based on Vision Foundation Model](https://mlsanity.com/paper/2403.10127/) (2024)
- [MMLSv2: A Multimodal Dataset for Martian Landslide Detection in Remote Sensing Imagery](https://mlsanity.com/paper/2602.08112/) (2026)

## Built on by (most cited first, 1 of 1)

- [Global-Local Contextual Progressive Expansion Network for Martian Landslide Segmentation in Multimodal Remote Sensing Imagery](https://mlsanity.com/paper/2609.13332/) (2026)

## Abstract (arXiv)

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.