Unsupervised distance learning by reciprocal kNN distance for image retrieval
Nenhuma Miniatura disponível
Data
2014-01-01
Autores
Pedronette, Daniel C. G. [UNESP]
Penatti, Otávio A. B.
Calumby, Rodrigo T.
Da S. Torres, Ricardo
Título da Revista
ISSN da Revista
Título de Volume
Editor
Resumo
This paper presents a novel unsupervised learning approach that takes into account the intrinsic dataset structure, which is represented in terms of the reciprocal neighborhood references found in different ranked lists. The proposed Reciprocal kNN Distance defines a more effective distance between two images, and is used to improve the effectiveness of image retrieval systems. Several experiments were conducted for different image retrieval tasks involving shape, color, and texture descriptors. The proposed approach is also evaluated on multimodal retrieval tasks, considering visual and textual descriptors. Experimental results demonstrate the effectiveness of proposed approach. The Reciprocal kNN Distance yields better results in terms of effectiveness than various state-of-the-art algorithms. Copyright © 2014 ACM.
Descrição
Palavras-chave
Content-based image retrieval, Unsupervised distance learning
Como citar
ICMR 2014 - Proceedings of the ACM International Conference on Multimedia Retrieval 2014, p. 345-352.