A new approach to sparse image representation using MMV and K-SVD

Authors

Jie Yang, University of WollongongFollow
Abdesselam Bouzerdoum, University of WollongongFollow
Son Lam Phung, University of WollongongFollow

RIS ID

31186

Publication Details

Yang, J., Bouzerdoum, A. & Phung, S. (2009). A new approach to sparse image representation using MMV and K-SVD. In J. Blanc-Talon, W. Philips, D. Popescu & P. Scheunders (Eds.), Advanced concepts for intelligent vision systems : ACIVS 2009 (pp. 200-209). Berlin, Germany: Springer-Verlag.

Abstract

This paper addresses the problem of image representation based on a sparse decomposition over a learned dictionary. We propose an improved matching pursuit algorithm for Multiple Measurement Vectors (MMV) and an adaptive algorithm for dictionary learning based on multi-Singular Value Decomposition (SVD), and combine them for image representation. Compared with the traditional K-SVD and orthogonal matching pursuit MMV (OMPMMV) methods, the proposed method runs faster and achieves a higher overall reconstruction accuracy.