Reproduce “NatureFusion: Rethinking Infrared and Visible Image Fusion as Natural Complementary Enhancement” 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

Jianfeng Ding, Linfeng Tang, Hao Zhang, Han Xu, Xunpeng Yi, Yibing Zhang, Meiqi Gong, Zhongyuan Wang and 1 more

NatureFusion generates visible-style images from registered visible–infrared pairs, using infrared cues to locate people and visible context to guide their appearance. It trains a generative fusion adapter from filtered demonstrations, then adds a task adapter that strengthens infrared’s contribution. On LLVIP-250, it recalls 47.1% of pedestrians missed by seven visible-image detectors, a defined missed set rather than all pedestrians.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.