egenioussBench: A New Dataset for Geospatial Visual Localisation

2026-05-06

AI Summary

This work addresses the lack of realistic, georeferenced, and scalable benchmarks for large-scale visual localization by introducing the first benchmark that integrates city-scale aerial 3D meshes with CityGML Level-of-Details 2 (LoD2) models. Leveraging centimeter-accurate smartphone images as queries, the benchmark features a non-co-visible test set and a sequential validation set. It incorporates depth-rendering-based co-visibility estimation, maximum independent set selection for query placement, and a multi-threshold binning evaluation protocol. The released dataset comprises 42 non-co-visible test images and 412 validation images, accompanied by a public leaderboard and evaluation code to enable fair, reproducible assessment of cross-view and cross-domain localization methods, thereby better reflecting real-world deployment challenges.

Abstract

We present egenioussBench, a visual localisation benchmark built on geospatial reference data: a city-scale airborne 3D mesh and a CityGML LoD2 model. This pairing reflects deployable mapping assets and supports true scalability beyond traditional SfM-based approaches. The query data comprise smartphone images with centimetre-accurate, map-independent ground truth obtained via PPK and GCP/CP-aided adjustment. From 2,709 images, we derive a non-co-visible subset by estimating the full co-visibility matrix from rendered depth and selecting a maximum independent set; the released data include a test split of 42 non-co-visible images with withheld ground truth and a validation split of 412 sequential images with poses, e.g. for training of pose regressors and self-validation. The benchmark features a public leaderboard evaluated with binning metrics at multiple pose-error thresholds alongside global statistics (median, RMSE, outlier ratio), ensuring fair, like-for-like comparison across mesh- and LoD2-based methods. Together, these design choices expose realistic cross-view and cross-domain challenges while providing a rigorous, scalable path for advancing large-scale visual localisation. We make the evaluation code and data availeable at https://github.com/fratopa/egenioussBench and https://www.egeniouss.eu/

Problem

• visual localisation
• geospatial reference data
• cross-view challenge
• cross-domain challenge
• scalable benchmark

Innovation

• visual localisation
• geospatial benchmark
• non-co-visible images
• CityGML LoD2
• scalable evaluation