Location Sensitive Image Retrieval and Tagging

Raul Gomez*, Jaume Gibert, Lluis Gomez, Dimosthenis Karatzas

*Autor corresponent d’aquest treball

Producció científica: Capítol de llibreCapítolRecercaAvaluat per experts

2 Cites (Scopus)

Resum

People from different parts of the globe describe objects and concepts in distinct manners. Visual appearance can thus vary across different geographic locations, which makes location a relevant contextual information when analysing visual data. In this work, we address the task of image retrieval related to a given tag conditioned on a certain location on Earth. We present LocSens, a model that learns to rank triplets of images, tags and coordinates by plausibility, and two training strategies to balance the location influence in the final ranking. LocSens learns to fuse textual and location information of multimodal queries to retrieve related images at different levels of location granularity, and successfully utilizes location information to improve image tagging.

Idioma originalAnglès
Títol de la publicacióComputer Vision – ECCV 2020 - 16th European Conference, Proceedings
EditorsAndrea Vedaldi, Horst Bischof, Thomas Brox, Jan-Michael Frahm
EditorSpringer Science and Business Media Deutschland GmbH
Pàgines649-665
Nombre de pàgines17
ISBN (imprès)9783030585167
DOIs
Estat de la publicacióPublicada - 2020

Sèrie de publicacions

NomLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volum12361 LNCS
ISSN (imprès)0302-9743
ISSN (electrònic)1611-3349

Fingerprint

Navegar pels temes de recerca de 'Location Sensitive Image Retrieval and Tagging'. Junts formen un fingerprint únic.

Com citar-ho