PaperScope
LIVE · 2026-09-03 05:40 UTC

AutoCompass: Accurate Visual Localization on Public Maps by Learning from Weak Labels

Javier Tirado-Garín, Alan Savio Paul, Shuai Chen, Axel Barroso-Laguna, Tommaso Cavallari, Daniyar Turmukhambetov, Victor Adrian Prisacariu, Eric Brachmann

Latestcs.CLcs.LGcs.AIcs.CV
arXiv ID
2609.02798 v1
Category
Submitted
2026-09-02

Abstract

Neural map matchers estimate an image's 3-DoF pose relative to a 2D map. These models are trained on large-scale datasets of geo-referenced images, whose position and heading labels often contain noise that affects the trained models. To address this, we present AutoCompass, a supervision approach for training neural map matchers from inaccurate absolute pose labels. First, we show that heading labels are unnecessary: trained from raw GPS labels, models learn to predict accurate headings, automatically. Second, defining a tolerance region around raw GPS improves positional accuracy. Third, if available, our supervision uses relative poses between training images, obtained via SLAM or SfM, which provide a more accurate training signal. Across driving and egocentric benchmarks, AutoCompass consistently outperforms counterparts trained with the usual strong reliance on absolute pose labels.

Comment: ECCV 2026

arXiv abs page · PDF · same-day batch