Problem
Research questions and friction points this paper is trying to address.
change detection
remote sensing
boundary delineation
region localization
binary change detection
MambaRefine-CD: MambaVision with Region-Boundary Temporal Refinement
🤖 AI Summary
Binary change detection in remote sensing struggles to simultaneously preserve the completeness of changed regions and the precision of their boundaries. This work proposes a region-boundary temporal refinement framework that leverages a shared MambaVision encoder to generate dual-stream features—region-aware and Sobel-conditioned boundary features—via a D-RBI module. The framework integrates CRAM-lite enhancement and an adaptive receptive field FPN decoder, and introduces a signed temporal evidence-based mechanism for decoupled region-boundary modeling along with a bounded residual refinement strategy to jointly optimize coarse predictions and fine boundaries. Evaluated on DSIFN-CD and WHU-CD benchmarks, the method achieves state-of-the-art F1 scores and IoU, while ablation studies confirm the effectiveness of its core components.
📝 Abstract
Binary change detection in remote sensing requires both complete changed-region localization and accurate boundary delineation. We present MambaRefine-CD, a region-boundary temporal refinement framework built on a shared MambaVision encoder. The proposed D-RBI module constructs temporal evidence from paired features, absolute differences, and signed differences, then separates it into region and Sobel-conditioned boundary streams. Region features are enhanced with CRAM-lite and decoded by an adaptive receptive-field FPN, while the finest boundary stream guides a bounded residual refinement of the coarse prediction. Experiments on DSIFN-CD and WHU-CD show strong changed-class F1 and IoU under verified evaluation settings, and ablations support the contribution of signed temporal evidence and the full region-boundary refinement pipeline.
Innovation
Methods, ideas, or system contributions that make the work stand out.
MambaVision
region-boundary refinement
signed temporal evidence
binary change detection
adaptive receptive-field FPN
🔎 Similar Papers
No similar papers found.
💼 Related Jobs
