TY - JOUR
T1 - Image recovery via transform learning and low-rank modeling
T2 - The power of complementary regularizers
AU - Wen, Bihan
AU - Li, Yanjun
AU - Bresler, Yoram
N1 - Funding Information:
Manuscript received August 3, 2018; revised July 15, 2019 and August 24, 2019; accepted August 25, 2019. Date of publication March 19, 2020; date of current version April 1, 2020. This work was supported in part by the National Science Foundation (NSF) under Grant CCF-1320953 and Grant IIS 14-47879. Bihan Wen was supported in part by the Ministry of Education, Singapore, through a start-up grant. The associate editor coordinating the review of this manuscript and approving it for publication was Dr. Denis Kouame. (Corresponding author: Bihan Wen.) Bihan Wen is with the School of Electrical and Electronic Engineering, Nanyang Technological University, Singapore 639798 (e-mail: bihan.wen@ntu.edu.sg).
Publisher Copyright:
© 2020 IEEE.
PY - 2020
Y1 - 2020
N2 - Recent works on adaptive sparse and on low-rank signal modeling have demonstrated their usefulness in various image/video processing applications. Patch-based methods exploit local patch sparsity, whereas other works apply low-rankness of grouped patches to exploit image non-local structures. However, using either approach alone usually limits performance in image reconstruction or recovery applications. In this work, we propose a simultaneous sparsity and low-rank model, dubbed STROLLR, to better represent natural images. In order to fully utilize both the local and non-local image properties, we develop an image restoration framework using a transform learning scheme with joint low-rank regularization. The approach owes some of its computational efficiency and good performance to the use of transform learning for adaptive sparse representation rather than the popular synthesis dictionary learning algorithms, which involve approximation of NP-hard sparse coding and expensive learning steps. We demonstrate the proposed framework in various applications to image denoising, inpainting, and compressed sensing based magnetic resonance imaging. Results show promising performance compared to state-of-the-art competing methods.
AB - Recent works on adaptive sparse and on low-rank signal modeling have demonstrated their usefulness in various image/video processing applications. Patch-based methods exploit local patch sparsity, whereas other works apply low-rankness of grouped patches to exploit image non-local structures. However, using either approach alone usually limits performance in image reconstruction or recovery applications. In this work, we propose a simultaneous sparsity and low-rank model, dubbed STROLLR, to better represent natural images. In order to fully utilize both the local and non-local image properties, we develop an image restoration framework using a transform learning scheme with joint low-rank regularization. The approach owes some of its computational efficiency and good performance to the use of transform learning for adaptive sparse representation rather than the popular synthesis dictionary learning algorithms, which involve approximation of NP-hard sparse coding and expensive learning steps. We demonstrate the proposed framework in various applications to image denoising, inpainting, and compressed sensing based magnetic resonance imaging. Results show promising performance compared to state-of-the-art competing methods.
KW - Block matching
KW - Collaborative filtering
KW - Image denoising
KW - Image inpainting
KW - Image reconstruction
KW - Machine learning
KW - Sparse representation
UR - http://www.scopus.com/inward/record.url?scp=85082833623&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85082833623&partnerID=8YFLogxK
U2 - 10.1109/TIP.2020.2980753
DO - 10.1109/TIP.2020.2980753
M3 - Article
C2 - 32203020
AN - SCOPUS:85082833623
SN - 1057-7149
VL - 29
SP - 5310
EP - 5323
JO - IEEE Transactions on Image Processing
JF - IEEE Transactions on Image Processing
M1 - 9042815
ER -