Benchmarking result diversification in social image retrieval

Ionescu, Bogdan (LAPI, University Politehnica of Bucharest, Romania / DISI, University of Trento, Italy) ; Popescu, Adrian (CEA, LIST, France) ; Müller, Henning (University of Applied Sciences and Arts Western Switzerland (HES-SO Valais-Wallis)) ; Menendez, Maria (DISI, University of Trento, Italy) ; Radu, Anca-Livia (LAPI, University Politehnica of Bucharest, Romania / DISI, University of Trento, Italy)

This article addresses the issue of retrieval result diversifica- tion in the context of social image retrieval and discusses the results achieved during the MediaEval 2013 benchmarking. 38 runs and their results are described and analyzed in this text. A comparison of the use of expert vs. crowdsourcing annotations shows that crowdsourcing results are slightly dif- ferent and have higher inter observer differences but results are comparable at lower cost. Multimodal approaches have best results in terms of cluster recall. Manual approaches can lead to high precision but often lower diversity. With this de- tailed results analysis we give future insights on this matter.


Mots-clés:
Type de conférence:
full paper
Faculté:
Economie et Services
Ecole:
HEG VS HES-SO Valais-Wallis - Haute Ecole de Gestion & Tourisme
Institut:
Institut Informatique de gestion
Classification:
Informatique
Adresse bibliogr.:
Paris, 27-30 October 2014
Date:
Paris
27-30 October 2014
2014
Pagination:
5 p.
Titre du document hôte:
Proceedings of IEEE International Conference on Image Processing 2014
Le document apparaît dans:



 Notice créée le 2015-09-04, modifiée le 2018-02-15

Fichiers:
Télécharger le document
PDF

Évaluer ce document:

Rate this document:
1
2
3
 
(Pas encore évalué)