Product detail

CrossLocate: Cross-Modal Large-Scale Visual Geo-Localization in Natural Environments using Rendered Modalities

TOMEŠEK, J. ČADÍK, M. BREJCHA, J.

Product type

software

Abstract

CrossLocate software implements the CrossLocate research paper published at the WACV 2022 conference. CrossLocate is a deep-learning method performing visual geo-localization, i.e. determining the geographic origin of photographs, which can be captured in natural environments, such as mountains. With this software, it is possible to train the localization method and evaluate it on a provided dataset. For more information and additional materials, please visit the project website: http://cphoto.fit.vutbr.cz/crosslocate/

Keywords

visual geo-localization, cross-modal, large-scale, image retrieval, natural environments, rendered modalities, deep learning

Create date

1. 1. 2022

Location

https://github.com/JanTomesek/CrossLocate

Possibilities of use

K využití výsledku jiným subjektem je vždy nutné nabytí licence

Licence fee

Poskytovatel licence na výsledek nepožaduje licenční poplatek

www

https://github.com/JanTomesek/CrossLocate