Approximate infinite-dimensional Region Covariance Descriptors for image classification

Masoud Faraki, Mehrtash T. Harandi, Fatih Porikli

Research output: Chapter in Book/Report/Conference proceedingConference PaperResearchpeer-review

26 Citations (Scopus)


We introduce methods to estimate infinite-dimensional Region Covariance Descriptors (RCovDs) by exploiting two feature mappings, namely random Fourier features and the Nyström method. In general, infinite-dimensional RCovDs offer better discriminatory power over their low-dimensional counterparts. However, the underlying Riemannian structure, i.e., the manifold of Symmetric Positive Definite (SPD) matrices, is out of reach to great extent for infinite-dimensional RCovDs. To overcome this difficulty, we propose to approximate the infinite-dimensional RCovDs by making use of the aforementioned explicit mappings. We will empirically show that the proposed finite-dimensional approximations of infinite-dimensional RCovDs consistently outperform the low-dimensional RCovDs for image classification task, while enjoying the Riemannian structure of the SPD manifolds. Moreover, our methods achieve the state-of-the-art performance on three different image classification tasks.

Original languageEnglish
Title of host publication2015 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2015)
EditorsDoug Gray, Doug Cochran
Place of PublicationPiscataway NJ USA
PublisherIEEE, Institute of Electrical and Electronics Engineers
Number of pages5
ISBN (Print)9781467369985
Publication statusPublished - 1 Jan 2015
Externally publishedYes
EventIEEE International Conference on Acoustics, Speech and Signal Processing 2015 - Brisbane Convention & Exhibition Centre, Brisbane, Australia
Duration: 19 Apr 201524 Apr 2015 (Proceedings)


ConferenceIEEE International Conference on Acoustics, Speech and Signal Processing 2015
Abbreviated titleICASSP 2015
Internet address


  • Region Covariance Descriptor
  • Reproducing Kernel Hilbert Space
  • Riemannian Geometry

Cite this