Combining visual dictionary, kernel-based similarity and learning strategy for image category retrieval

Abstract : This paper presents a search engine architecture, RETIN, aiming at retrieving com- plex categories in large image databases. For indexing, a scheme based on a two-step quantization process is presented to compute visual codebooks. The similarity be- tween images is represented in a kernel framework. Such a similarity is combined with online learning strategies motivated by recent Machine-Learning developments such as Active Learning. Additionally, an offine supervised learning is embedded in the kernel framework, offering a real opportunity to learn semantic categories. Experiments with real scenario carried out from the Corel Photo database demon- strate the effciency and the relevance of the RETIN strategy and its outstanding performances in comparison to up-to-date strategies.
Document type :
Journal articles
Complete list of metadatas

Cited literature [51 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-00520290
Contributor : Philippe-Henri Gosselin <>
Submitted on : Wednesday, September 22, 2010 - 6:41:26 PM
Last modification on : Friday, October 4, 2019 - 12:14:02 PM
Long-term archiving on : Thursday, December 23, 2010 - 3:16:13 AM

File

gosselin08cviu.pdf
Publisher files allowed on an open archive

Identifiers

  • HAL Id : hal-00520290, version 1

Citation

Philippe-Henri Gosselin, Matthieu Cord, Sylvie Philipp-Foliguet. Combining visual dictionary, kernel-based similarity and learning strategy for image category retrieval. Computer Vision and Image Understanding, Elsevier, 2008, 110 (3), pp.403-417. ⟨hal-00520290⟩

Share

Metrics

Record views

274

Files downloads

465