Building-Guided Pseudo-Label Learning for Cross-Modal Building Damage Mapping

TL;DR

This paper introduces a novel pseudo-label learning framework leveraging building priors and multi-modal data to improve cross-modal building damage mapping, achieving state-of-the-art results in a major competition.

Contribution

It proposes a building-guided pseudo-label refinement strategy that enhances damage classification accuracy in cross-modal remote sensing images.

Findings

Achieved highest mIoU score of 54.28% in the IEEE GRSS Data Fusion Contest.

Secured first place in the competition with the proposed method.

Demonstrated effectiveness of building priors in reducing pseudo-label uncertainty.

Abstract

Accurate building damage assessment using bi-temporal multi-modal remote sensing images is essential for effective disaster response and recovery planning. This study proposes a novel Building-Guided Pseudo-Label Learning Framework to address the challenges of mapping building damage from pre-disaster optical and post-disaster SAR images. First, we train a series of building extraction models using pre-disaster optical images and building labels. To enhance building segmentation, we employ multi-model fusion and test-time augmentation strategies to generate pseudo-probabilities, followed by a low-uncertainty pseudo-label training method for further refinement. Next, a change detection model is trained on bi-temporal cross-modal images and damaged building labels. To improve damage classification accuracy, we introduce a building-guided low-uncertainty pseudo-label refinement strategy, which leverages building priors from the previous step to guide pseudo-label generation for damaged buildings, reducing uncertainty and enhancing reliability. Experimental results on the 2025 IEEE GRSS Data Fusion Contest dataset demonstrate the effectiveness of our approach, which achieved the highest mIoU score (54.28%) and secured first place in the competition.