Highly overcomplete sparse coding

Bruno A. Olshausen

doi:10.1117/12.2013504

14 March 2013 Highly overcomplete sparse coding

Bruno A. Olshausen

Proceedings Volume 8651, Human Vision and Electronic Imaging XVIII; 86510S (2013) https://doi.org/10.1117/12.2013504
Event: IS&T/SPIE Electronic Imaging, 2013, Burlingame, California, United States

Abstract

This paper explores sparse coding of natural images in the highly overcomplete regime. We show that as the overcompleteness ratio approaches l0x, new types of dictionary elements emerge beyond the classical Gabor function shape obtained from complete or only modestly overcomplete sparse coding. These more diverse dic tionaries allow images to be approximated with lower L1 norm (for a fixed SNR), and the coefficients exhibit steeper decay. We also evaluate the learned dictionaries in a denoising task, showing that higher degrees of overcompleteness yield modest gains in peformance.

Citation Download Citation

Bruno A. Olshausen "Highly overcomplete sparse coding", Proc. SPIE 8651, Human Vision and Electronic Imaging XVIII, 86510S (14 March 2013); https://doi.org/10.1117/12.2013504

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available