Reproduce “MatchCityLoc: Aerial Localization in LoD City Models via Surface Normal Matching” 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

Oussema Dhaouadi, Zuria Bauer, Johannes Michael Meier, Marc Pollefeys, Daniel Cremers, Olaf Wysocki

MatchCityLoc addresses drone localization in cities where GNSS may be unreliable, matching surface normals between camera images and textureless Level-of-Detail city models to estimate pose from a prior. On SemCityLockeD with LoD3 models, it reaches 70.33% joint recall at the 2 m / 2° threshold, versus 49.63% for an adapted silhouette baseline, with lower median translation error; it requires a close prior and is not real-time.

Summary by alphaXiv.

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