Answers are grounded in this paper and cite their sources.
Accurate building change detection in high-resolution remote sensing imagery requires models that can simultaneously capture fine-grained geometric detail and semantically meaningful temporal differences. Existing deep learning approaches predominantly rely on convolutional or transformer-based architectures with uniform tokenization and global prediction strategies, often overlooking object-level spatial structure and instance-aware alignment. In this paper, we propose RADNet, a novel Region-Aware Dual-path network for building change detection, that introduces an Instance-Consistent Tokenizer (ICT) which leverages segmentation masks to generate structured region tokens, preserving building-level semantics. These tokens are fused through a Global-Local Attention Encoder (GLAE) that captures multi-scale contextual dependencies and cross-temporal interactions. To enhance prediction fidelity, we introduce a Cross-Scale Correspondence Decoder (CSCD) that aligns token-level differences via correlation maps and reconstructs dense change maps with instance-level precision. Extensive experiments on EGY-BCD, LEVIR-CD, and SYSU-CD datasets demonstrate that RADNet outperforms state-of-the-art baselines, achieving superior boundary accuracy and region consistency, with average gains of +2.15% IoU and +1.71% F1.
No takes yet. Share an insight, caveat, or question.
Bianca-Cerasela-Zelia Blaga (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: