Map-free Visual Relocalization: Metric Pose Relative to a Single Image
Abstract
![Map-free Visual Relocation Paper Thumbnail](/mapfree-reloc-benchmark/paper-thumbnail.png)
Dataset
We make available a dataset of 655 places, collected by non-expert users world-wide. The full dataset is publicly available here.
![Animated image showing example scenes from the dataset.](mapfree-reloc-benchmark/zoomed.gif)
Benchmark
We propose new metrics based on the reprojection error of virtual objects. To successfully compete in our benchmark, methods need to be precise but also provide a robust confidence estimate of their predictions. You can submit to our public leaderboard here.
BibTeX
Please cite our paper if you use our dataset or code.
@inproceedings{arnold2022mapfree,
title={Map-free Visual Relocalization: Metric Pose Relative to a Single Image},
author={Arnold, Eduardo and Wynn, Jamie and Vicente, Sara and Garcia-Hernando, Guillermo and Monszpart, {\'{A}}ron and Prisacariu, Victor Adrian and Turmukhambetov, Daniyar and Brachmann, Eric},
booktitle={ECCV},
year={2022},
}