Learning to Align Semantic Segmentation and 2.5D Maps for Geolocalization

Anil Armagan, Martin Hirzer, Peter M. Roth, Vincent Lepetit

Publikation: Beitrag in Buch/Bericht/KonferenzbandBeitrag in einem KonferenzbandBegutachtung

Abstract

We present an efficient method for geolocalization in urban environments starting from a coarse estimate of the location provided by a GPS and using a simple untextured 2.5D model of the surrounding buildings. Our key contribution is a novel efficient and robust method to optimize the pose: We train a Deep Network to predict the best direction to improve a pose estimate, given a semantic segmentation of the input image and a rendering of the buildings from this estimate. We then iteratively apply this CNN until converging to a good pose. This approach avoids the use of reference images of the surroundings, which are difficult to acquire and match, while 2.5D models are broadly available. We can therefore apply it to places unseen during training.
Originalspracheenglisch
TitelProceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
PublikationsstatusVeröffentlicht - 2017
Veranstaltung2017 IEEE Conference on Computer Vision and Pattern Recognition: CVPR 2017 - Honolulu, USA / Vereinigte Staaten
Dauer: 21 Juli 201726 Juli 2017

Konferenz

Konferenz2017 IEEE Conference on Computer Vision and Pattern Recognition
KurztitelCVPR 2017
Land/GebietUSA / Vereinigte Staaten
OrtHonolulu
Zeitraum21/07/1726/07/17

Fingerprint

Untersuchen Sie die Forschungsthemen von „Learning to Align Semantic Segmentation and 2.5D Maps for Geolocalization“. Zusammen bilden sie einen einzigartigen Fingerprint.

Dieses zitieren