Reproduce “CoDeRNet: Selective Cross-Task Routing under Heterogeneous Supervision for Change Detection and Captioning” 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
Eunki Cho, Hyeon Bae Kim, Seong Tae Kim
CoDeRNet combines satellite image change detection and captioning through separate branches that exchange features using learned estimates of recipient need and sender reliability. On LEVIR-MCI, it scores 141.79 CIDEr-D, compared with 140.29 for MCINet. This reference-caption consensus metric does not directly measure factual truth; results also show competitive localization, while performance varies across datasets and metrics.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.