To Learn or Not to Learn: Visual Localization from Essential Matrices
Paper i proceeding, 2020

Visual localization is the problem of estimating a camera within a scene and a key technology for autonomous robots. State-of-the-art approaches for accurate visual localization use scene-specific representations, resulting in the overhead of constructing these models when applying the techniques to new scenes. Recently, learned approaches based on relative pose estimation have been proposed, carrying the promise of easily adapting to new scenes. However, they are currently significantly less accurate than state-of-the-art approaches. In this paper, we are interested in analyzing this behavior. To this end, we propose a novel framework for visual localization from relative poses. Using a classical feature-based approach within this framework, we show state-of-the-art performance. Replacing the classical approach with learned alternatives at various levels, we then identify the reasons for why deep learned approaches do not perform well. Based on our analysis, we make recommendations for future work.


Qunjie Zhou

Technische Universität München

Torsten Sattler

Chalmers, Elektroteknik, Signalbehandling och medicinsk teknik

Marc Pollefeys

Eidgenössische Technische Hochschule Zürich (ETH)

Laura Leal-Taixe

Technische Universität München

Proceedings - IEEE International Conference on Robotics and Automation

10504729 (ISSN)

3319-3326 9196607
9781728173955 (ISBN)

2020 IEEE International Conference on Robotics and Automation, ICRA 2020
Paris, France,


Robotteknik och automation

Datavetenskap (datalogi)

Datorseende och robotik (autonoma system)



Mer information

Senast uppdaterat