TY - GEN
T1 - Boosted dictionaries for image restoration based on sparse representations
AU - Natesan Ramamurthy, Karthikeyan
AU - Thiagarajan, Jayaraman J.
AU - Spanias, Andreas
AU - Sattigeri, Prasanna
PY - 2013/10/18
Y1 - 2013/10/18
N2 - Sparse representations using learned dictionaries have been successful in several image processing applications. However, using a single dictionary model in inverse problems may lead to instability in estimation. In this paper, we propose to perform image restoration using an ensemble of weak dictionaries that incorporate prior knowledge about the form of linear corruption. The dictionary learned in each round of the training procedure is optimized for the training examples having high reconstruction error in the previous round. The weak dictionaries are either obtained using a weighted K-Means or an example-selection approach. The final restored data is computed as a convex combination of data restored in individual rounds. Results with compressed recovery of standard images show that the proposed dictionaries result in a better performance compared to using a single dictionary obtained with a traditional alternating minimization approach.
AB - Sparse representations using learned dictionaries have been successful in several image processing applications. However, using a single dictionary model in inverse problems may lead to instability in estimation. In this paper, we propose to perform image restoration using an ensemble of weak dictionaries that incorporate prior knowledge about the form of linear corruption. The dictionary learned in each round of the training procedure is optimized for the training examples having high reconstruction error in the previous round. The weak dictionaries are either obtained using a weighted K-Means or an example-selection approach. The final restored data is computed as a convex combination of data restored in individual rounds. Results with compressed recovery of standard images show that the proposed dictionaries result in a better performance compared to using a single dictionary obtained with a traditional alternating minimization approach.
KW - Boosting
KW - Dictionary learning
KW - Image restoration
KW - Sparse representations
UR - http://www.scopus.com/inward/record.url?scp=84890500577&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84890500577&partnerID=8YFLogxK
U2 - 10.1109/ICASSP.2013.6637918
DO - 10.1109/ICASSP.2013.6637918
M3 - Conference contribution
AN - SCOPUS:84890500577
SN - 9781479903566
T3 - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
SP - 1583
EP - 1587
BT - 2013 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2013 - Proceedings
T2 - 2013 38th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2013
Y2 - 26 May 2013 through 31 May 2013
ER -