Graph-based selective rank fusion for unsupervised 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
Resumo
Nowadays, there is a great variety of visual features available for image retrieval tasks. While fusion strategies have been established as a promising alternative, an inherent difficulty in unsupervised scenarios is the task of selecting the features to combine. In this paper, a Graph-based Selective Rank Fusion is proposed. The graph is used to represent the effectiveness estimation of features and the complementarity among them. The selected combinations are defined by the Connected Components of the graph. High-effective retrieval results were achieved through a comprehensive experimental evaluation considering different public datasets, dozens of features and comparisons with related methods. Relative gains up to +54.73% were obtained in relation to the best isolated feature.
Descrição
Palavras-chave
Content-based image retrieval, Correlation measure, Effectiveness estimation, Rank-aggregation, Unsupervised late fusion
Idioma
Inglês
Citação
Pattern Recognition Letters, v. 135, p. 82-89.
