Semi-Supervised Building Change Detection From Bitemporal Remote Sensing Images Leveraging Visual–Language Models and Consistency Learning

Building change detection from bi-temporal remote sensing images aims to identify newly constructed, demolished, or modified buildings by comparing observations acquired at different times. A major limitation of existing supervised approaches is their strong dependence on pixel-wise building and change annotations, which are costly and labor-intensive to obtain at scale. To address this challenge, we propose SemiBCD, a semi-supervised framework that integrates visual–language model (VLM) priors with consistency learning for building change detection under limited supervision. A VLM pre-trained on natural images is first employed to generate building pseudo-labels for remote sensing imagery. To mitigate domain shift and suppress noise in these pseudo-labels, we introduce an Uncertainty-Aware Pseudo-Label Refinement (UAPLR) module that progressively improves pseudo-label reliability during training. Leveraging a small set of annotated change masks together with the refined pseudo-labels, we further design a dual consistency learning strategy, combining perturbation-based multi-view consistency and temporal-hierarchical alignment to better exploit unlabeled data. Experiments on the LEVIR-CD and WHU-CD benchmarks, as well as the WHU Building dataset, demonstrate that SemiBCD consistently outperforms representative semi-supervised baselines under low-label regimes. The results indicate that incorporating VLM priors with uncertainty-aware refinement and consistency regularization provides an effective solution for reducing annotation dependency in building change detection. The source code will be released upon publication at https://github.com/ hejianglin2001/SemiBCD.git.