Search papers, labs, and topics across Lattice.
The paper introduces WILD-SAM, a parameter-efficient fine-tuning framework that adapts the Segment Anything Model (SAM) for landslide detection in wrapped InSAR interferograms. It addresses the spectral domain shift between natural images and interferometric phase data by integrating a Phase-Aware Mixture-of-Experts (PA-MoE) Adapter into the frozen encoder and a Wavelet-Guided Subband Enhancement (WGSE) strategy to generate frequency-aware dense prompts. Experiments on ISSLIDE and ISSLIDE+ benchmarks show that WILD-SAM achieves state-of-the-art performance in target completeness and contour fidelity.
SAM's impressive segmentation abilities don't directly translate to scientific imaging, but a novel adaptation strategy unlocks its potential for high-precision landslide detection from noisy radar data.
Detecting slow-moving landslides directly from wrapped Interferometric Synthetic Aperture Radar (InSAR) interferograms is crucial for efficient geohazard monitoring, yet it remains fundamentally challenged by severe phase ambiguity and complex coherence noise. While the Segment Anything Model (SAM) offers a powerful foundation for segmentation, its direct transfer to wrapped phase data is hindered by a profound spectral domain shift, which suppresses the high-frequency fringes essential for boundary delineation. To bridge this gap, we propose WILD-SAM, a novel parameter-efficient fine-tuning framework specifically designed to adapt SAM for high-precision landslide detection on wrapped interferograms. Specifically, the architecture integrates a Phase-Aware Mixture-of-Experts (PA-MoE) Adapter into the frozen encoder to align spectral distributions and introduces a Wavelet-Guided Subband Enhancement (WGSE) strategy to generate frequency-aware dense prompts. The PA-MoE Adapter exploits a dynamic routing mechanism across heterogeneous convolutional experts to adaptively aggregate multi-scale spectral-textural priors, effectively aligning the distribution discrepancy between natural images and interferometric phase data. Meanwhile, the WGSE strategy leverages discrete wavelet transforms to explicitly disentangle high-frequency subbands and refine directional phase textures, injecting these structural cues as dense prompts to ensure topological integrity along sharp landslide boundaries. Extensive experiments on the ISSLIDE and ISSLIDE+ benchmarks demonstrate that WILD-SAM achieves state-of-the-art performance, significantly outperforming existing methods in both target completeness and contour fidelity.
