A kernel-based active learning strategy for content-based image retrieval - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2010

A kernel-based active learning strategy for content-based image retrieval

Résumé

Active learning methods have attracted many researchers in the content-based image retrieval (CBIR) community. In this paper, we propose an efficient kernel-based active learning strategy to improve the retrieval performance of CBIR systems using class probability distributions. The proposed method learns for each class a nonlinear kernel which transforms the original feature space into a more effective one. The distances between user’s request and database images are then learned and computed in the kernel space. Experimental results show that the proposed kernel-based active learning approach not only improves the retrieval performances of kernel distance without learning, but also outperforms other kernel metric learning methods
Fichier non déposé

Dates et versions

hal-01381537 , version 1 (14-10-2016)

Identifiants

Citer

Imane Daoudi, Khalid Idrissi. A kernel-based active learning strategy for content-based image retrieval. CBMI2010, 8th International Workshop on content-Based Multimedia Indexing, Jun 2010, Grenoble, France. pp.1-6, ⟨10.1109/CBMI.2010.5529915⟩. ⟨hal-01381537⟩
82 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More