Tensor sparse coding for positive definite matrices

Research output: Contribution to journalArticlepeer-review

32 Scopus citations


In recent years, there has been extensive research on sparse representation of vector-valued signals. In the matrix case, the data points are merely vectorized and treated as vectors thereafter (for example, image patches). However, this approach cannot be used for all matrices, as it may destroy the inherent structure of the data. Symmetric positive definite (SPD) matrices constitute one such class of signals, where their implicit structure of positive eigenvalues is lost upon vectorization. This paper proposes a novel sparse coding technique for positive definite matrices, which respects the structure of the Riemannian manifold and preserves the positivity of their eigenvalues, without resorting to vectorization. Synthetic and real-world computer vision experiments with region covariance descriptors demonstrate the need for and the applicability of the new sparse coding model. This work serves to bridge the gap between the sparse modeling paradigm and the space of positive definite matrices.

Original languageEnglish (US)
Article number6574845
Pages (from-to)592-605
Number of pages14
JournalIEEE Transactions on Pattern Analysis and Machine Intelligence
Issue number3
StatePublished - Mar 2014


  • Sparse coding
  • computer vision
  • optimization
  • positive definite matrices
  • region covariance descriptors


Dive into the research topics of 'Tensor sparse coding for positive definite matrices'. Together they form a unique fingerprint.

Cite this