Reproduce “TERDNet: Transformer Encoder-Recurrent Decoder Network for Scene Change Detection” 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
Jiae Yoon, Ue-Hwan Kim
TERDNet detects pixel-level changes between two images of the same place by combining transformer features from multiple layers with local correlation information and repeatedly refining a change mask in a recurrent decoder. On four benchmarks, it scores higher than the listed prior methods; on VL-CMU-CD, it reaches 83.4% F1 versus 80.2% for C-3PO, without demonstrating improved robot navigation.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.