Abstract
We show how to use simple 2.5D maps of buildings and recent advances in image segmentation and machine learning to geo-localize an input image of an urban scene: We first extract the façades of the buildings and their edges from the image, and then look for the orientation and location that align a 3D rendering of the map with these segments. We discuss how to use a 3D tracking system to acquire the data required for training the segmentation method, the segmentation itself, and how we use the segmentations to evaluate the quality of the alignment.
Original language | English |
---|---|
Title of host publication | Proceedings of the Joint Urban Remote Sensing Event (JURSE) |
Publication status | Published - 2017 |
Event | Joint Urban Remote Sensing Event 2017 - Dubai, United Arab Emirates Duration: 6 Mar 2017 → 8 Mar 2017 |
Conference
Conference | Joint Urban Remote Sensing Event 2017 |
---|---|
Abbreviated title | JURSE 2017 |
Country/Territory | United Arab Emirates |
City | Dubai |
Period | 6/03/17 → 8/03/17 |
Fingerprint
Dive into the research topics of 'Semantic Segmentation for 3D Localization in Urban Environments'. Together they form a unique fingerprint.Prizes
-
Best Paper Award at JURSE 2017
Armagan, Anil (Recipient), Hirzer, Martin (Recipient) & Lepetit, Vincent (Recipient), 8 Mar 2017
Prize: Prizes / Medals / Awards