Reproduce “STHN: Deep Homography Estimation for UAV Thermal Geo-localization with Satellite Imagery” 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
Jiuhong Xiao, Ning Zhang, Daniel Tortei, Giuseppe Loianno
STHN localizes a night-flying UAV by aligning thermal imagery with a satellite map when GNSS is unreliable. Its two-stage method estimates alignment over the map, crops the predicted region, and refines it, avoiding dense retrieval. On Boson-nighttime at a 512-meter center-distance limit, it reports 12.12-meter center error and 63.9-millisecond latency, versus 13.52-meter and 4,918.9-millisecond latency for retrieval, under the benchmark protocol.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.