Maximum unfolded embedding: Formulation, solution, and application for image clustering

Huan Wang, Shuicheng Yan, Thomas S Huang, Xiaoou Tang

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In this paper, we present a novel spectral analysis algorithm for image clustering. First, the image manifold is embedded onto a low-dimensional feature space with dual objectives, i.e., maximizing the distances of faraway sample pairs meanwhile preserving the local manifold structure, which essentially results in a Trace Ratio optimization problem. Then an efficient iterative procedure is proposed to directly optimize the trace ratio and finally the clustering process is implemented on the derived low-dimensional embedding. Moreover, the linear approximation is also presented for handling the out-of-sample data. Experimental results show that our algorithm, referred to as Maximum Unfolded Embedding, brings an encouraging improvement in clustering accuracy over the state-of-the-art algorithms, such as K-Means, PCA-Kmeans, normalized cut [8], and Locality Preserving Clustering [13].

Original languageEnglish (US)
Title of host publicationProceedings of the 14th Annual ACM International Conference on Multimedia, MM 2006
Pages45-48
Number of pages4
DOIs
StatePublished - Dec 1 2006
Event14th Annual ACM International Conference on Multimedia, MM 2006 - Santa Barbara, CA, United States
Duration: Oct 23 2006Oct 27 2006

Publication series

NameProceedings of the 14th Annual ACM International Conference on Multimedia, MM 2006

Other

Other14th Annual ACM International Conference on Multimedia, MM 2006
Country/TerritoryUnited States
CitySanta Barbara, CA
Period10/23/0610/27/06

Keywords

  • Image clustering
  • Maximum unfolded embedding
  • Spectral analysis

ASJC Scopus subject areas

  • Computer Science(all)
  • Media Technology

Fingerprint

Dive into the research topics of 'Maximum unfolded embedding: Formulation, solution, and application for image clustering'. Together they form a unique fingerprint.

Cite this