LM-Reloc: Levenberg-Marquardt Based Direct Visual Relocalization

Lukas Von Stumberg, Patrick Wenzel, Nan Yang, Daniel Cremers

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

14 Scopus citations

Abstract

We present LM-Reloc-a novel approach for visual relocalization based on direct image alignment. In contrast to prior works that tackle the problem with a feature-based formulation, the proposed method does not rely on feature matching and RANSAC. Hence, the method can utilize not only corners but any region of the image with gradients. In particular, we propose a loss formulation inspired by the classical Levenberg-Marquardt algorithm to train LM-Net. The learned features significantly improve the robustness of direct image alignment, especially for relocalization across different conditions. To further improve the robustness of LM-Net against large image baselines, we propose a pose estimation network, CorrPoseNet, which regresses the relative pose to bootstrap the direct image alignment. Evaluations on the CARLA and Oxford RobotCar relocalization tracking benchmark show that our approach delivers more accurate results than previous state-of-the-art methods while being comparable in terms of robustness.

Original languageEnglish
Title of host publicationProceedings - 2020 International Conference on 3D Vision, 3DV 2020
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages968-977
Number of pages10
ISBN (Electronic)9781728181288
DOIs
StatePublished - Nov 2020
Event8th International Conference on 3D Vision, 3DV 2020 - Virtual, Fukuoka, Japan
Duration: 25 Nov 202028 Nov 2020

Publication series

NameProceedings - 2020 International Conference on 3D Vision, 3DV 2020

Conference

Conference8th International Conference on 3D Vision, 3DV 2020
Country/TerritoryJapan
CityVirtual, Fukuoka
Period25/11/2028/11/20

Fingerprint

Dive into the research topics of 'LM-Reloc: Levenberg-Marquardt Based Direct Visual Relocalization'. Together they form a unique fingerprint.

Cite this