Ecole d'ingénieur et centre de recherche en Sciences du numérique

Natural language access to video databases

Francis, Danny; Pidou, Paul; Merialdo, Bernard; Huet, Benoit

BIGMM 2017, 3rd International Conference on Multimedia Big Data, 19-21 April 2017, Laguna Hills, USA

This paper deals with natural language access to video databases. Two approaches are proposed: in the first one we use queries to find images similar to  ideo keyframes, and in the second one we generate text descriptions from keyframes and compare them with queries. We propose four implementations of these approaches: one implementation of the first approach, two implementations of the second one and one implementation mixing both approaches. The results of our implementations are discussed, in particular regarding the visual content of natural language queries.

Document Doi Bibtex

Titre:Natural language access to video databases
Type:Conférence
Langue:English
Ville:Laguna Hills
Pays:ÉTATS-UNIS
Date:
Département:Data Science
Eurecom ref:5199
Copyright: © 2017 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE.
Bibtex: @inproceedings{EURECOM+5199, doi = {http://dx.doi.org/10.1109/BigMM.2017.34}, year = {2017}, title = {{N}atural language access to video databases}, author = {{F}rancis, {D}anny and {P}idou, {P}aul and {M}erialdo, {B}ernard and {H}uet, {B}enoit }, booktitle = {{BIGMM} 2017, 3rd {I}nternational {C}onference on {M}ultimedia {B}ig {D}ata, 19-21 {A}pril 2017, {L}aguna {H}ills, {USA} }, address = {{L}aguna {H}ills, {\'{E}}{TATS}-{UNIS}}, month = {04}, url = {http://www.eurecom.fr/publication/5199} }
Voir aussi: