Abstract
The recent years have witnessed significant advances in image deblurring. In general, the success of deblurring methods depends heavily on extraction of salient structures from a blurry image for kernel estimation. Most deblurring methods often operate on high-resolution images where contours or edges can be extracted for kernel estimation. However, recovering reliable structures from low-resolution images becomes extremely challenging. In this paper, we propose a spatially variant deblurring algorithm for low-resolution images based on the exemplars. To exploit the exemplar information, we develop a super-resolution guided method to help the restoration of reliable image structures which can be used for kernel estimation. Experimental evaluations against the state-of-the-art methods show that the proposed algorithm performs favorably in deblurring low-resolution images. Furthermore, we show that the SR results obtained as byproducts in our method are comparable compared to other blind SR methods.
| Original language | English |
|---|---|
| Title of host publication | Computer Vision - ACCV 2016 Workshops - ACCV 2016 International Workshops, Revised Selected Papers |
| Editors | Jiwen Lu, Kai-Kuang Ma, Chu-Song Chen |
| Publisher | Springer Verlag |
| Pages | 111-127 |
| Number of pages | 17 |
| ISBN (Print) | 9783319544069 |
| DOIs | |
| Publication status | Published - 2017 Jan 1 |
| Event | 13th Asian Conference on Computer Vision, ACCV 2016 - Taipei, Taiwan, Province of China Duration: 2016 Nov 20 → 2016 Nov 24 |
Publication series
| Name | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) |
|---|---|
| Volume | 10116 LNCS |
| ISSN (Print) | 0302-9743 |
| ISSN (Electronic) | 1611-3349 |
Other
| Other | 13th Asian Conference on Computer Vision, ACCV 2016 |
|---|---|
| Country | Taiwan, Province of China |
| City | Taipei |
| Period | 16/11/20 → 16/11/24 |
Fingerprint
All Science Journal Classification (ASJC) codes
- Theoretical Computer Science
- Computer Science(all)
Cite this
}
Debluring low-resolution images. / Pan, Jinshan; Hu, Zhe; Su, Zhixun; Yang, Ming Hsuan.
Computer Vision - ACCV 2016 Workshops - ACCV 2016 International Workshops, Revised Selected Papers. ed. / Jiwen Lu; Kai-Kuang Ma; Chu-Song Chen. Springer Verlag, 2017. p. 111-127 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10116 LNCS).Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
TY - GEN
T1 - Debluring low-resolution images
AU - Pan, Jinshan
AU - Hu, Zhe
AU - Su, Zhixun
AU - Yang, Ming Hsuan
PY - 2017/1/1
Y1 - 2017/1/1
N2 - The recent years have witnessed significant advances in image deblurring. In general, the success of deblurring methods depends heavily on extraction of salient structures from a blurry image for kernel estimation. Most deblurring methods often operate on high-resolution images where contours or edges can be extracted for kernel estimation. However, recovering reliable structures from low-resolution images becomes extremely challenging. In this paper, we propose a spatially variant deblurring algorithm for low-resolution images based on the exemplars. To exploit the exemplar information, we develop a super-resolution guided method to help the restoration of reliable image structures which can be used for kernel estimation. Experimental evaluations against the state-of-the-art methods show that the proposed algorithm performs favorably in deblurring low-resolution images. Furthermore, we show that the SR results obtained as byproducts in our method are comparable compared to other blind SR methods.
AB - The recent years have witnessed significant advances in image deblurring. In general, the success of deblurring methods depends heavily on extraction of salient structures from a blurry image for kernel estimation. Most deblurring methods often operate on high-resolution images where contours or edges can be extracted for kernel estimation. However, recovering reliable structures from low-resolution images becomes extremely challenging. In this paper, we propose a spatially variant deblurring algorithm for low-resolution images based on the exemplars. To exploit the exemplar information, we develop a super-resolution guided method to help the restoration of reliable image structures which can be used for kernel estimation. Experimental evaluations against the state-of-the-art methods show that the proposed algorithm performs favorably in deblurring low-resolution images. Furthermore, we show that the SR results obtained as byproducts in our method are comparable compared to other blind SR methods.
UR - http://www.scopus.com/inward/record.url?scp=85016170335&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85016170335&partnerID=8YFLogxK
U2 - 10.1007/978-3-319-54407-6_8
DO - 10.1007/978-3-319-54407-6_8
M3 - Conference contribution
AN - SCOPUS:85016170335
SN - 9783319544069
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 111
EP - 127
BT - Computer Vision - ACCV 2016 Workshops - ACCV 2016 International Workshops, Revised Selected Papers
A2 - Lu, Jiwen
A2 - Ma, Kai-Kuang
A2 - Chen, Chu-Song
PB - Springer Verlag
ER -