×

Learning optimal representations for image retrieval applications. (English) Zbl 1029.68799

Bakker, Erwin M. (ed.) et al., Image and video retrieval. Second international conference, CIVR 2003, Urbana-Champaign, IL, USA, July 24-25, 2003. Proceedings. Berlin: Springer. Lect. Notes Comput. Sci. 2728, 50-60 (2003).
Summary: This paper presents an MCMC stochastic gradient algorithm for finding representations with optimal retrieval performance on given image datasets. For linear subspaces in the image space and the spectral space, the problem is formulated as that of optimization on a Grassmann manifold. By exploiting the underlying geometry of the manifold, a computationally effective algorithm is developed. The feasibility and effectiveness of the proposed algorithm are demonstrated through extensive experimental results.
For the entire collection see [Zbl 1026.68831].

MSC:

68U99 Computing methodologies and applications
68P20 Information storage and retrieval of data
68T45 Machine vision and scene understanding
68U35 Computing methodologies for information systems (hypertext navigation, interfaces, decision support, etc.)
PDFBibTeX XMLCite
Full Text: Link