Denoising with greedy-like pursuit algorithms

Raja Giryes*, Michael Elad

*Corresponding author for this work

Research output: Contribution to journalConference articlepeer-review

2 Scopus citations


This paper provides theoretical guarantees for denoising performance of greedy-like methods. Those include Compressive Sampling Matching Pursuit (CoSaMP), Subspace Pursuit (SP), and Iterative Hard Thresholding (IHT). Our results show that the denoising obtained with these algorithms is a constant and a log-factor away from the oracle's performance, if the signal's representation is sufficiently sparse. Turning to practice, we show how to convert these algorithms to work without knowing the target cardinality, and instead constrain the solution to an error-budget. Denoising tests on synthetic data and image patches show the potential in this stagewise technique as a replacement of the classical OMP.

Original languageEnglish
Pages (from-to)1475-1479
Number of pages5
JournalEuropean Signal Processing Conference
StatePublished - 2011
Externally publishedYes
Event19th European Signal Processing Conference, EUSIPCO 2011 - Barcelona, Spain
Duration: 29 Aug 20112 Sep 2011


Dive into the research topics of 'Denoising with greedy-like pursuit algorithms'. Together they form a unique fingerprint.

Cite this