Hybrid linear modeling via local best-fit flats

Teng Zhang, Arthur Szlam, Yi Wang, Gilad Lerman

Research output: Contribution to journalArticle

116 Scopus citations

Abstract

We present a simple and fast geometric method for modeling data by a union of affine subspaces. The method begins by forming a collection of local best-fit affine subspaces, i.e., subspaces approximating the data in local neighborhoods. The correct sizes of the local neighborhoods are determined automatically by the Jones' β 2 numbers (we prove under certain geometric conditions that our method finds the optimal local neighborhoods). The collection of subspaces is further processed by a greedy selection procedure or a spectral method to generate the final model. We discuss applications to tracking-based motion segmentation and clustering of faces under different illuminating conditions. We give extensive experimental evidence demonstrating the state of the art accuracy and speed of the suggested algorithms on these problems and also on synthetic hybrid linear data as well as the MNIST handwritten digits data; and we demonstrate how to use our algorithms for fast determination of the number of affine subspaces.

Original languageEnglish (US)
Pages (from-to)217-240
Number of pages24
JournalInternational Journal of Computer Vision
Volume100
Issue number3
DOIs
StatePublished - Dec 1 2012

Keywords

  • Face clustering
  • High-dimensional data
  • Hybrid linear modeling
  • Local PCA
  • Motion segmentation
  • Spectral clustering
  • Subspace clustering

ASJC Scopus subject areas

  • Software
  • Computer Vision and Pattern Recognition
  • Artificial Intelligence

Fingerprint Dive into the research topics of 'Hybrid linear modeling via local best-fit flats'. Together they form a unique fingerprint.

  • Cite this