Cluster-Wise Ratio Tests for Fast Camera Localization

Raul Diaz, Charless C. Fowlkes; Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 2017, pp. 19-28


Feature point matching for camera localization suffers from scalability problems. Even when feature descriptors associated with 3D scene points are locally unique, as coverage grows, similar or repeated features become increasingly common. As a result, the standard distance ratio-test used to identify reliable image feature points is overly restrictive and rejects many good candidate matches. We propose a simple coarse-to-fine strategy that uses conservative approximations to robust local ratio-tests that can be computed efficiently using global approximate k-nearest neighbor search. We treat these forward matches as votes in camera pose space and use them to prioritize back-matching within candidate camera pose clusters, exploiting feature co-visibility captured by the 3D model camera pose graph. This approach achieves state-of-the-art camera pose estimation results on a variety of benchmarks, outperforming several methods that use more complicated data structures and that make more restrictive assumptions on camera pose. We carry out diagnostic analyses on a difficult test dataset containing globally repetitive structure which suggest our approach successfully adapts to the challenges of large-scale pose estimation.

Related Material

[pdf] [arXiv]
author = {Diaz, Raul and Fowlkes, Charless C.},
title = {Cluster-Wise Ratio Tests for Fast Camera Localization},
booktitle = {Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Workshops},
month = {July},
year = {2017}