Publicação: Unsupervised rank diffusion for content-based image retrieval
Carregando...
Data
Orientador
Coorientador
Pós-graduação
Curso de graduação
Título da Revista
ISSN da Revista
Título de Volume
Editor
Tipo
Artigo
Direito de acesso
Acesso restrito
Resumo
Despite the continuous development of features and mid-level representations, effectively and reliably measuring the similarity among images remains a challenging problem in image retrieval tasks. Once traditional measures consider only pairwise analysis, context-sensitive measures capable of exploiting the intrinsic manifold structure became indispensable for improving the retrieval performance. In this scenario, diffusion processes and rank-based methods are the most representative approaches. This paper proposes a novel hybrid method, named rank diffusion, which uses a diffusion process based on ranking information. The proposed method consists in a diffusion-based re-ranking approach, which propagates contextual information through a diffusion process defined in terms of top-ranked objects, reducing the computational complexity of the proposed algorithm. Extensive experiments considering a rigorous experimental protocol were conducted on six public image datasets and several different descriptors. Experimental results and comparison with state-of-the-art methods demonstrate that high effectiveness gains can be obtained, despite the low-complexity of the algorithm proposed.
Descrição
Palavras-chave
Content-based image retrieval, Rank diffusion, Unsupervised learning
Idioma
Inglês
Como citar
Neurocomputing, v. 260, p. 478-489.
