Reproduce “MambaRefine-CD: MambaVision with Region-Boundary Temporal Refinement” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Dineth Perera, Thaariq Firdous, Oshadha Samarakoon, Roshan Godaliyadda, Parakrama Ekanayake, Vijitha Herath
MambaRefine-CD detects changed regions in pairs of remote-sensing images while refining their boundaries. A shared MambaVision encoder feeds separate region and boundary streams; signed temporal differences contribute evidence, and a bounded residual adjusts the coarse prediction. On DSIFN-CD, the full pipeline raises BF1 from 61.36% to 71.94% against a MambaVision-FPN baseline, supporting the combined pipeline, not the residual head alone.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.