Learning dual-pixel alignment for defocus deblurring

It is a challenging task to recover sharp image from a single defocus blurry image in real-world applications. On many modern cameras, dual-pixel (DP) sensors create two-image views, based on which stereo information can be exploited to benefit defocus deblurring. Despite the impressive results achi...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Neurocomputing (Amsterdam) Ročník 616; s. 128880
Hlavní autoři: Li, Yu, Yi, Yaling, Shu, Xinya, Ren, Dongwei, Li, Qince, Zuo, Wangmeng
Médium: Journal Article
Jazyk:angličtina
Vydáno: Elsevier B.V 01.02.2025
Témata:
ISSN:0925-2312
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Abstract It is a challenging task to recover sharp image from a single defocus blurry image in real-world applications. On many modern cameras, dual-pixel (DP) sensors create two-image views, based on which stereo information can be exploited to benefit defocus deblurring. Despite the impressive results achieved by existing DP defocus deblurring methods, the misalignment between DP image views is still not studied, leaving room for improving DP defocus deblurring. In this work, we propose a Dual-Pixel Alignment Network (DPANet) for defocus deblurring. Generally, DPANet is an encoder–decoder with skip-connections, where two branches with shared parameters in the encoder are employed to extract and align deep features from left and right views, and one decoder is adopted to fuse aligned features for predicting the sharp image. Due to that DP views suffer from different blur amounts, it is not trivial to align left and right views. To this end, we propose novel encoder alignment module (EAM) and decoder alignment module (DAM). In particular, a correlation layer is suggested in EAM to measure the disparity between DP views, whose deep features can then be accordingly aligned using deformable convolutions. DAM can further enhance the alignment of skip-connected features from encoder and deep features in decoder. By introducing several EAMs and DAMs, DP views in DPANet can be well aligned for better predicting latent sharp image. Experimental results on real-world datasets show that our DPANet is notably superior to state-of-the-art deblurring methods in reducing defocus blur while recovering visually plausible sharp structures and textures.
AbstractList It is a challenging task to recover sharp image from a single defocus blurry image in real-world applications. On many modern cameras, dual-pixel (DP) sensors create two-image views, based on which stereo information can be exploited to benefit defocus deblurring. Despite the impressive results achieved by existing DP defocus deblurring methods, the misalignment between DP image views is still not studied, leaving room for improving DP defocus deblurring. In this work, we propose a Dual-Pixel Alignment Network (DPANet) for defocus deblurring. Generally, DPANet is an encoder–decoder with skip-connections, where two branches with shared parameters in the encoder are employed to extract and align deep features from left and right views, and one decoder is adopted to fuse aligned features for predicting the sharp image. Due to that DP views suffer from different blur amounts, it is not trivial to align left and right views. To this end, we propose novel encoder alignment module (EAM) and decoder alignment module (DAM). In particular, a correlation layer is suggested in EAM to measure the disparity between DP views, whose deep features can then be accordingly aligned using deformable convolutions. DAM can further enhance the alignment of skip-connected features from encoder and deep features in decoder. By introducing several EAMs and DAMs, DP views in DPANet can be well aligned for better predicting latent sharp image. Experimental results on real-world datasets show that our DPANet is notably superior to state-of-the-art deblurring methods in reducing defocus blur while recovering visually plausible sharp structures and textures.
ArticleNumber 128880
Author Li, Yu
Ren, Dongwei
Zuo, Wangmeng
Shu, Xinya
Li, Qince
Yi, Yaling
Author_xml – sequence: 1
  givenname: Yu
  surname: Li
  fullname: Li, Yu
  email: liyuhit@outlook.com
– sequence: 2
  givenname: Yaling
  surname: Yi
  fullname: Yi, Yaling
  email: csylyi@outlook.com
– sequence: 3
  givenname: Xinya
  surname: Shu
  fullname: Shu, Xinya
  email: shuxinyahit@outlook.com
– sequence: 4
  givenname: Dongwei
  surname: Ren
  fullname: Ren, Dongwei
  email: rendongweihit@gmail.com
– sequence: 5
  givenname: Qince
  surname: Li
  fullname: Li, Qince
  email: qinceli@hit.edu.cn
– sequence: 6
  givenname: Wangmeng
  surname: Zuo
  fullname: Zuo, Wangmeng
  email: wmzuo@hit.edu.cn
BookMark eNp9j71OwzAUhT0UibbwBgx5gQTf68RJFiRU8SdVYoHZcu3rylHqVHaC4O1JFWams5zz6XwbtgpDIMbugBfAQd53RaDJDKcCOZYFYNM0fMXWvMUqRwF4zTYpdZxDDdiuWbknHYMPx8xOus_P_pv6TPf-GE4UxswNMbPkBjOlOQ_9FOPcvWFXTveJbv9yyz6fnz52r_n-_eVt97jPDVbVmFsoa066MlVNspHiYBspAdA4xMaVFsmQE62VYJ2sHQqHUpRwaCst0GkhtqxcuCYOKUVy6hz9SccfBVxdbFWnFlt1sVWL7Tx7WGY0f_vyFFUynoIh6yOZUdnB_w_4BbsyY44
Cites_doi 10.1007/s11063-024-11455-w
10.1109/TIP.2021.3055613
10.1109/CVPR52729.2023.00570
10.1109/CVPR.2017.295
10.1109/CVPR52688.2022.00564
10.1109/ICCV48922.2021.00229
10.1109/CVPR46437.2021.01458
10.1109/TIP.2020.3002345
10.1109/CVPR52729.2023.00557
10.1109/ICCV51070.2023.01158
10.1109/CVPR.2015.7298665
10.1109/CVPR.2018.00068
10.1109/CVPR.2019.01250
10.1109/TIP.2021.3113185
10.1109/TIP.2016.2526907
10.1109/CVPR42600.2020.00230
10.1109/TIFS.2020.3035879
10.1109/CVPR.2017.179
10.1109/CVPR.2019.00165
10.1109/TIP.2016.2528042
10.1109/ICCV.2015.123
10.1007/978-3-030-01267-0_32
10.1109/TIP.2020.3045630
10.1109/TIP.2003.819861
10.1109/JAS.2022.105563
10.1109/CVPR.2018.00931
10.1109/ICCV.2015.316
10.4236/jcc.2013.16003
10.1364/OL.399204
10.1109/TIP.2018.2881830
10.1109/CVPR46437.2021.00432
10.1109/TIP.2021.3061901
10.1109/CVPRW.2019.00247
10.1007/s11063-021-10604-9
10.1109/ICCV.2019.00772
10.1109/TIP.2017.2771563
10.3390/s150305747
10.1364/JOSAA.12.000058
10.1109/JPROC.2023.3238524
10.1109/CVPR.2019.01125
10.1109/TII.2018.2884211
10.1109/CVPR.2019.00613
10.1109/ICCV.2019.00897
10.1109/CVPR.2019.00953
10.1109/TIP.2016.2617460
ContentType Journal Article
Copyright 2024 Elsevier B.V.
Copyright_xml – notice: 2024 Elsevier B.V.
DBID AAYXX
CITATION
DOI 10.1016/j.neucom.2024.128880
DatabaseName CrossRef
DatabaseTitle CrossRef
DatabaseTitleList
DeliveryMethod fulltext_linktorsrc
Discipline Computer Science
ExternalDocumentID 10_1016_j_neucom_2024_128880
S0925231224016515
GroupedDBID ---
--K
--M
.DC
.~1
0R~
123
1B1
1~.
1~5
4.4
457
4G.
53G
5VS
7-5
71M
8P~
9JM
9JN
AABNK
AACTN
AAEDT
AAEDW
AAIKJ
AAKOC
AALRI
AAOAW
AAQFI
AAXKI
AAXLA
AAXUO
AAYFN
ABBOA
ABCQJ
ABFNM
ABJNI
ABMAC
ACDAQ
ACGFS
ACRLP
ACZNC
ADBBV
ADEZE
AEBSH
AEKER
AENEX
AFJKZ
AFKWA
AFTJW
AFXIZ
AGHFR
AGUBO
AGWIK
AGYEJ
AHHHB
AHZHX
AIALX
AIEXJ
AIKHN
AITUG
AJOXV
AKRWK
ALMA_UNASSIGNED_HOLDINGS
AMFUW
AMRAJ
AOUOD
AXJTR
BKOJK
BLXMC
CS3
DU5
EBS
EFJIC
EO8
EO9
EP2
EP3
F5P
FDB
FIRID
FNPLU
FYGXN
G-Q
GBLVA
GBOLZ
IHE
J1W
KOM
MO0
MOBAO
N9A
O-L
O9-
OAUVE
OZT
P-8
P-9
P2P
PC.
Q38
ROL
RPZ
SDF
SDG
SDP
SES
SEW
SPC
SPCBC
SSN
SSV
SSZ
T5K
ZMT
~G-
29N
9DU
AAQXK
AATTM
AAYWO
AAYXX
ABWVN
ABXDB
ACLOT
ACNNM
ACRPL
ACVFH
ADCNI
ADJOM
ADMUD
ADNMO
AEIPS
AEUPX
AFPUW
AGQPQ
AIGII
AIIUN
AKBMS
AKYEP
ANKPU
APXCP
ASPBG
AVWKF
AZFZN
CITATION
EFKBS
EFLBG
EJD
FEDTE
FGOYB
HLZ
HVGLF
HZ~
LG9
M41
R2-
SBC
WUQ
XPP
~HD
ID FETCH-LOGICAL-c255t-d1470ea5c57e6863bd866112cf228f4d2ecef39d61df67f23f26341b95a32fa33
ISICitedReferencesCount 1
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001363744800001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 0925-2312
IngestDate Sat Nov 29 06:34:02 EST 2025
Sat Dec 14 16:14:39 EST 2024
IsPeerReviewed true
IsScholarly true
Keywords Image deblurring
Defocus deblurring
Language English
LinkModel OpenURL
MergedId FETCHMERGED-LOGICAL-c255t-d1470ea5c57e6863bd866112cf228f4d2ecef39d61df67f23f26341b95a32fa33
ParticipantIDs crossref_primary_10_1016_j_neucom_2024_128880
elsevier_sciencedirect_doi_10_1016_j_neucom_2024_128880
PublicationCentury 2000
PublicationDate 2025-02-01
2025-02-00
PublicationDateYYYYMMDD 2025-02-01
PublicationDate_xml – month: 02
  year: 2025
  text: 2025-02-01
  day: 01
PublicationDecade 2020
PublicationTitle Neurocomputing (Amsterdam)
PublicationYear 2025
Publisher Elsevier B.V
Publisher_xml – sequence: 0
  name: Elsevier B.V
References S.W. Zamir, A. Arora, S. Khan, M. Hayat, F.S. Khan, M.-H. Yang, L. Shao, Multi-stage progressive image restoration, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2021, pp. 14821–14831.
A. Abuolaim, M. Delbracio, D. Kelly, M.S. Brown, P. Milanfar, Learning to reduce defocus blur by realistically modeling dual-pixel data, in: Proceedings of the IEEE/CVF International Conference on Computer Vision, 2021, pp. 2289–2298.
Paszke, Gross, Massa, Lerer, Bradbury, Chanan, Killeen, Lin, Gimelshein, Antiga others (b51) 2019; 32
Śliwiński, Wachel (b17) 2013; 1
Hong, Yu, Zhang, Jin, Lee (b4) 2018; 15
Jung, Kim, Jang, Ha, Sohn (b36) 2021; 30
Punnappurath, Abuolaim, Afifi, Brown (b42) 2020
X. Zhu, H. Hu, S. Lin, J. Dai, Deformable convnets v2: More deformable, better results, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR, 2019, pp. 9308–9316.
Mao, Shen, Yang (b32) 2016
Zou, Chen, Shi, Guo, Ye (b5) 2023; 111
Wu, Hong, Zhang, He (b7) 2021; 53
D. Sun, X. Yang, M.-Y. Liu, J. Kautz, Pwc-net: Cnns for optical flow using pyramid, warping, and cost volume, in: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2018, pp. 8934–8943.
D’Andrés, Salvador, Kochale, Süsstrunk (b10) 2016; 25
Charbonnier, Blanc-Feraud, Aubert, Barlaud (b46) 1994; vol. 2
Krishnan, Fergus (b14) 2009; 22
S. Zhou, J. Zhang, W. Zuo, H. Xie, J. Pan, J. Ren, DAVANet: Stereo deblurring with view aggregation, in: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, CVPR, 2019, pp. 10996–11005.
Abuolaim, Brown (b26) 2020
A. Dosovitskiy, P. Fischer, E. Ilg, P. Hausser, C. Hazirbas, V. Golkov, P. Van Der Smagt, D. Cremers, T. Brox, Flownet: Learning optical flow with convolutional networks, in: Proceedings of the IEEE International Conference on Computer Vision, 2015, pp. 2758–2766.
Yi, Eramian (b9) 2016; 25
X. Wang, K.C. Chan, K. Yu, C. Dong, C. Change Loy, Edvr: Video restoration with enhanced deformable convolutional networks, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 2019.
Liu, Zhou, Liao (b39) 2016; 25
Ronneberger, Fischer, Brox (b31) 2015
C. Herrmann, R.S. Bowen, N. Wadhwa, R. Garg, Q. He, J.T. Barron, R. Zabih, Learning to autofocus, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR, p. 2020.
Kong, Sun, Liu, Jiang, Li, Shi (b2) 2020; 29
E. Ilg, N. Mayer, T. Saikia, M. Keuper, A. Dosovitskiy, T. Brox, Flownet 2.0: Evolution of optical flow estimation with deep networks, in: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2017, pp. 2462–2470.
J. Lee, S. Lee, S. Cho, S. Lee, Deep defocus map estimation using domain adaptation, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR, 2019, pp. 12222–12230.
R. Garg, N. Wadhwa, S. Ansari, J.T. Barron, Learning single camera depth estimation using dual-pixels, in: Proceedings of the IEEE/CVF International Conference on Computer Vision, 2019, pp. 7628–7637.
J. Shi, L. Xu, J. Jia, Just noticeable defocus blur detection and estimation, in: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, CVPR, 2015, pp. 657–665.
Liang, Jiang, Liu, Ma (b29) 2022; 9
Wu, Zhou, Liu, Ni, Fan (b23) 2020; 16
Fish, Brinicombe, Pike, Walker (b15) 1995; 12
L. Pan, S. Chowdhury, R. Hartley, M. Liu, H. Zhang, H. Li, Dual pixel exploration: Simultaneous depth estimation and image restoration, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR, 2021, pp. 4340–4349.
Wang, Bovik, Sheikh, Simoncelli (b54) 2004; 13
Y. Quan, X. Yao, H. Ji, Single image defocus deblurring via implicit neural inverse kernels, in: Proceedings of the IEEE/CVF International Conference on Computer Vision, 2023, pp. 12600–12610.
R. Zhang, P. Isola, A.A. Efros, E. Shechtman, O. Wang, The unreasonable effectiveness of deep features as a perceptual metric, in: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2018, pp. 586–595.
A. Abuolaim, A. Punnappurath, M.S. Brown, Revisiting autofocus for smartphone cameras, in: Proceedings of the European Conference on Computer Vision, ECCV, 2018, pp. 523–537.
Zeng, Wang, Mao, Liu, Peng, Chen (b37) 2018; 28
Kong, Dong, Yang, Pan (b41) 2024
Fan, Hong, Zeng, Liu (b6) 2024; 56
H. Zhang, Y. Dai, H. Li, P. Koniusz, Deep stacked hierarchical multi-patch network for image deblurring, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2019, pp. 5978–5986.
Guo, Feng, Gao, Liu, Wang (b1) 2021; 30
A. Punnappurath, M.S. Brown, Reflection removal using a dual-pixel sensor, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2019, pp. 1556–1565.
K. He, X. Zhang, S. Ren, J. Sun, Delving deep into rectifiers: Surpassing human-level performance on imagenet classification, in: Proceedings of the IEEE International Conference on Computer Vision, 2015, pp. 1026–1034.
J. Park, Y.-W. Tai, D. Cho, I. So Kweon, A unified approach of multi-scale deep and hand-crafted features for defocus estimation, in: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, CVPR, 2017, pp. 1736–1745.
Zhang, Sun (b38) 2021; 30
K. Dp, J. Ba, Adam: A method for stochastic optimization, in: Proc. of the 3rd International Conference for Learning Representations, ICLR, 2015.
Jang, Yoo, Kim, Paik (b16) 2015; 15
Karaali, Jung (b11) 2017; 27
Y. Quan, Z. Wu, H. Ji, Neumann network with recursive kernels for single image defocus deblurring, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2023, pp. 5754–5763.
Deng, Zhang, Zhong (b24) 2020; 45
Zhang, Wadhwa, Orts-Escolano, Häne, Fanello, Garg (b22) 2020
O. Kupyn, T. Martyniuk, J. Wu, Z. Wang, Deblurgan-v2: Deblurring (orders-of-magnitude) faster and better, in: Proceedings of the IEEE/CVF International Conference on Computer Vision, 2019, pp. 8878–8887.
Sun, Zheng, Lu (b3) 2021; 30
S.W. Zamir, A. Arora, S. Khan, M. Hayat, F.S. Khan, M.-H. Yang, Restormer: Efficient transformer for high-resolution image restoration, in: CVPR, 2022.
L. Kong, J. Dong, J. Ge, M. Li, J. Pan, Efficient frequency domain-based transformers for high-quality image deblurring, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2023, pp. 5886–5895.
10.1016/j.neucom.2024.128880_b47
10.1016/j.neucom.2024.128880_b48
Zhang (10.1016/j.neucom.2024.128880_b38) 2021; 30
10.1016/j.neucom.2024.128880_b45
Wu (10.1016/j.neucom.2024.128880_b23) 2020; 16
10.1016/j.neucom.2024.128880_b49
Liu (10.1016/j.neucom.2024.128880_b39) 2016; 25
Jung (10.1016/j.neucom.2024.128880_b36) 2021; 30
10.1016/j.neucom.2024.128880_b8
Fan (10.1016/j.neucom.2024.128880_b6) 2024; 56
10.1016/j.neucom.2024.128880_b50
Fish (10.1016/j.neucom.2024.128880_b15) 1995; 12
Paszke (10.1016/j.neucom.2024.128880_b51) 2019; 32
10.1016/j.neucom.2024.128880_b55
10.1016/j.neucom.2024.128880_b52
10.1016/j.neucom.2024.128880_b53
Wang (10.1016/j.neucom.2024.128880_b54) 2004; 13
10.1016/j.neucom.2024.128880_b12
10.1016/j.neucom.2024.128880_b13
Sun (10.1016/j.neucom.2024.128880_b3) 2021; 30
10.1016/j.neucom.2024.128880_b18
10.1016/j.neucom.2024.128880_b19
Charbonnier (10.1016/j.neucom.2024.128880_b46) 1994; vol. 2
Zhang (10.1016/j.neucom.2024.128880_b22) 2020
Ronneberger (10.1016/j.neucom.2024.128880_b31) 2015
Śliwiński (10.1016/j.neucom.2024.128880_b17) 2013; 1
10.1016/j.neucom.2024.128880_b21
Kong (10.1016/j.neucom.2024.128880_b41) 2024
10.1016/j.neucom.2024.128880_b20
Krishnan (10.1016/j.neucom.2024.128880_b14) 2009; 22
10.1016/j.neucom.2024.128880_b25
Liang (10.1016/j.neucom.2024.128880_b29) 2022; 9
Guo (10.1016/j.neucom.2024.128880_b1) 2021; 30
Hong (10.1016/j.neucom.2024.128880_b4) 2018; 15
10.1016/j.neucom.2024.128880_b27
10.1016/j.neucom.2024.128880_b28
D’Andrés (10.1016/j.neucom.2024.128880_b10) 2016; 25
Kong (10.1016/j.neucom.2024.128880_b2) 2020; 29
Yi (10.1016/j.neucom.2024.128880_b9) 2016; 25
Abuolaim (10.1016/j.neucom.2024.128880_b26) 2020
Mao (10.1016/j.neucom.2024.128880_b32) 2016
Punnappurath (10.1016/j.neucom.2024.128880_b42) 2020
Zou (10.1016/j.neucom.2024.128880_b5) 2023; 111
Karaali (10.1016/j.neucom.2024.128880_b11) 2017; 27
10.1016/j.neucom.2024.128880_b33
10.1016/j.neucom.2024.128880_b30
Zeng (10.1016/j.neucom.2024.128880_b37) 2018; 28
Deng (10.1016/j.neucom.2024.128880_b24) 2020; 45
10.1016/j.neucom.2024.128880_b34
10.1016/j.neucom.2024.128880_b35
Wu (10.1016/j.neucom.2024.128880_b7) 2021; 53
10.1016/j.neucom.2024.128880_b40
10.1016/j.neucom.2024.128880_b43
Jang (10.1016/j.neucom.2024.128880_b16) 2015; 15
10.1016/j.neucom.2024.128880_b44
References_xml – year: 2024
  ident: b41
  article-title: Efficient visual state space model for image deblurring
– volume: 30
  start-page: 3419
  year: 2021
  end-page: 3433
  ident: b38
  article-title: Joint depth and defocus estimation from a single image using physical consistency
  publication-title: IEEE Trans. Image Process.
– volume: 22
  start-page: 1033
  year: 2009
  end-page: 1041
  ident: b14
  article-title: Fast image deconvolution using hyper-laplacian priors
  publication-title: Adv. Neural Inf. Process. Syst.
– volume: 25
  start-page: 1626
  year: 2016
  end-page: 1638
  ident: b9
  article-title: Lbp-based segmentation of defocus blur
  publication-title: IEEE Trans. Image Process.
– volume: 25
  start-page: 1660
  year: 2016
  end-page: 1673
  ident: b10
  article-title: Non-parametric blur map regression for depth of field extension
  publication-title: IEEE Trans. Image Process.
– reference: Y. Quan, Z. Wu, H. Ji, Neumann network with recursive kernels for single image defocus deblurring, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2023, pp. 5754–5763.
– reference: K. He, X. Zhang, S. Ren, J. Sun, Delving deep into rectifiers: Surpassing human-level performance on imagenet classification, in: Proceedings of the IEEE International Conference on Computer Vision, 2015, pp. 1026–1034.
– reference: S.W. Zamir, A. Arora, S. Khan, M. Hayat, F.S. Khan, M.-H. Yang, Restormer: Efficient transformer for high-resolution image restoration, in: CVPR, 2022.
– volume: 16
  start-page: 1440
  year: 2020
  end-page: 1451
  ident: b23
  article-title: Single-shot face anti-spoofing for dual pixel camera
  publication-title: IEEE Trans. Inf. Forensics Secur.
– reference: Y. Quan, X. Yao, H. Ji, Single image defocus deblurring via implicit neural inverse kernels, in: Proceedings of the IEEE/CVF International Conference on Computer Vision, 2023, pp. 12600–12610.
– reference: K. Dp, J. Ba, Adam: A method for stochastic optimization, in: Proc. of the 3rd International Conference for Learning Representations, ICLR, 2015.
– volume: 30
  start-page: 1812
  year: 2021
  end-page: 1824
  ident: b1
  article-title: Exploring the effects of blur and deblurring to visual object tracking
  publication-title: IEEE Trans. Image Process.
– reference: J. Shi, L. Xu, J. Jia, Just noticeable defocus blur detection and estimation, in: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, CVPR, 2015, pp. 657–665.
– reference: S. Zhou, J. Zhang, W. Zuo, H. Xie, J. Pan, J. Ren, DAVANet: Stereo deblurring with view aggregation, in: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, CVPR, 2019, pp. 10996–11005.
– volume: 111
  start-page: 257
  year: 2023
  end-page: 276
  ident: b5
  article-title: Object detection in 20 years: A survey
  publication-title: Proc. IEEE
– volume: 13
  start-page: 600
  year: 2004
  end-page: 612
  ident: b54
  article-title: Image quality assessment: from error visibility to structural similarity
  publication-title: IEEE Trans. Image Process.
– volume: 56
  start-page: 27
  year: 2024
  ident: b6
  article-title: A deep convolutional encoder–decoder–restorer architecture for image deblurring
  publication-title: Neural Process. Lett.
– reference: A. Abuolaim, M. Delbracio, D. Kelly, M.S. Brown, P. Milanfar, Learning to reduce defocus blur by realistically modeling dual-pixel data, in: Proceedings of the IEEE/CVF International Conference on Computer Vision, 2021, pp. 2289–2298.
– volume: vol. 2
  start-page: 168
  year: 1994
  end-page: 172
  ident: b46
  article-title: Two deterministic half-quadratic regularization algorithms for computed imaging
  publication-title: Proceedings of 1st International Conference on Image Processing
– reference: J. Lee, S. Lee, S. Cho, S. Lee, Deep defocus map estimation using domain adaptation, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR, 2019, pp. 12222–12230.
– reference: L. Kong, J. Dong, J. Ge, M. Li, J. Pan, Efficient frequency domain-based transformers for high-quality image deblurring, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2023, pp. 5886–5895.
– reference: X. Wang, K.C. Chan, K. Yu, C. Dong, C. Change Loy, Edvr: Video restoration with enhanced deformable convolutional networks, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 2019.
– year: 2016
  ident: b32
  article-title: Image restoration using very deep convolutional encoder–decoder networks with symmetric skip connections
  publication-title: Advances in Neural Information Processing Systems
– volume: 25
  start-page: 5943
  year: 2016
  end-page: 5956
  ident: b39
  article-title: Defocus map estimation from a single image based on two-parameter defocus model
  publication-title: IEEE Trans. Image Process.
– reference: O. Kupyn, T. Martyniuk, J. Wu, Z. Wang, Deblurgan-v2: Deblurring (orders-of-magnitude) faster and better, in: Proceedings of the IEEE/CVF International Conference on Computer Vision, 2019, pp. 8878–8887.
– start-page: 234
  year: 2015
  end-page: 241
  ident: b31
  article-title: U-Net: convolutional networks for biomedical image segmentation
  publication-title: International Conference on Medical Image Computing and Computer-Assisted Intervention
– reference: A. Dosovitskiy, P. Fischer, E. Ilg, P. Hausser, C. Hazirbas, V. Golkov, P. Van Der Smagt, D. Cremers, T. Brox, Flownet: Learning optical flow with convolutional networks, in: Proceedings of the IEEE International Conference on Computer Vision, 2015, pp. 2758–2766.
– start-page: 582
  year: 2020
  end-page: 598
  ident: b22
  article-title: Du 2 net: Learning depth estimation from dual-cameras and dual-pixels
  publication-title: European Conference on Computer Vision
– reference: D. Sun, X. Yang, M.-Y. Liu, J. Kautz, Pwc-net: Cnns for optical flow using pyramid, warping, and cost volume, in: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2018, pp. 8934–8943.
– volume: 53
  start-page: 4419
  year: 2021
  end-page: 4436
  ident: b7
  article-title: Stack-based scale-recurrent network for face image deblurring
  publication-title: Neural Process. Lett.
– start-page: 111
  year: 2020
  end-page: 126
  ident: b26
  article-title: Defocus deblurring using dual-pixel data
  publication-title: European Conference on Computer Vision
– reference: R. Garg, N. Wadhwa, S. Ansari, J.T. Barron, Learning single camera depth estimation using dual-pixels, in: Proceedings of the IEEE/CVF International Conference on Computer Vision, 2019, pp. 7628–7637.
– volume: 15
  start-page: 5747
  year: 2015
  end-page: 5762
  ident: b16
  article-title: Sensor-based auto-focusing system using multi-scale feature extraction and phase correlation matching
  publication-title: Sensors
– reference: A. Abuolaim, A. Punnappurath, M.S. Brown, Revisiting autofocus for smartphone cameras, in: Proceedings of the European Conference on Computer Vision, ECCV, 2018, pp. 523–537.
– reference: J. Park, Y.-W. Tai, D. Cho, I. So Kweon, A unified approach of multi-scale deep and hand-crafted features for defocus estimation, in: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, CVPR, 2017, pp. 1736–1745.
– reference: H. Zhang, Y. Dai, H. Li, P. Koniusz, Deep stacked hierarchical multi-patch network for image deblurring, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2019, pp. 5978–5986.
– volume: 15
  start-page: 3952
  year: 2018
  end-page: 3961
  ident: b4
  article-title: Multimodal face-pose estimation with multitask manifold deep learning
  publication-title: IEEE Trans. Ind. Inf.
– volume: 32
  start-page: 8026
  year: 2019
  end-page: 8037
  ident: b51
  article-title: Pytorch: An imperative style, high-performance deep learning library
  publication-title: Adv. Neural Inf. Process. Syst.
– start-page: 1
  year: 2020
  end-page: 12
  ident: b42
  article-title: Modeling defocus-disparity in dual-pixel sensors
  publication-title: 2020 IEEE International Conference on Computational Photography
– reference: R. Zhang, P. Isola, A.A. Efros, E. Shechtman, O. Wang, The unreasonable effectiveness of deep features as a perceptual metric, in: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2018, pp. 586–595.
– reference: A. Punnappurath, M.S. Brown, Reflection removal using a dual-pixel sensor, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2019, pp. 1556–1565.
– volume: 30
  start-page: 8170
  year: 2021
  end-page: 8183
  ident: b36
  article-title: Multi-task learning framework for motion estimation and dynamic scene deblurring
  publication-title: IEEE Trans. Image Process.
– volume: 28
  start-page: 2107
  year: 2018
  end-page: 2115
  ident: b37
  article-title: A local metric for defocus blur detection based on cnn feature learning
  publication-title: IEEE Trans. Image Process.
– volume: 29
  start-page: 7389
  year: 2020
  end-page: 7398
  ident: b2
  article-title: Foveabox: Beyound anchor-based object detection
  publication-title: IEEE Trans. Image Process.
– volume: 9
  start-page: 878
  year: 2022
  end-page: 892
  ident: b29
  article-title: Bambnet: A blur-aware multi-branch network for dual-pixel defocus deblurring
  publication-title: IEEE/CAA J. Autom. Sin.
– reference: L. Pan, S. Chowdhury, R. Hartley, M. Liu, H. Zhang, H. Li, Dual pixel exploration: Simultaneous depth estimation and image restoration, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR, 2021, pp. 4340–4349.
– volume: 12
  start-page: 58
  year: 1995
  end-page: 65
  ident: b15
  article-title: Blind deconvolution by means of the richardson–lucy algorithm
  publication-title: J. Opt. Soc. Amer. A
– reference: X. Zhu, H. Hu, S. Lin, J. Dai, Deformable convnets v2: More deformable, better results, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR, 2019, pp. 9308–9316.
– reference: S.W. Zamir, A. Arora, S. Khan, M. Hayat, F.S. Khan, M.-H. Yang, L. Shao, Multi-stage progressive image restoration, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2021, pp. 14821–14831.
– volume: 1
  start-page: 11
  year: 2013
  ident: b17
  article-title: A simple model for on-sensor phase-detection autofocusing algorithm
  publication-title: J. Comput. Commun.
– reference: E. Ilg, N. Mayer, T. Saikia, M. Keuper, A. Dosovitskiy, T. Brox, Flownet 2.0: Evolution of optical flow estimation with deep networks, in: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2017, pp. 2462–2470.
– volume: 45
  start-page: 4734
  year: 2020
  end-page: 4737
  ident: b24
  article-title: Image-free real-time 3-d tracking of a fast-moving object using dual-pixel detection
  publication-title: Opt. Lett.
– volume: 30
  start-page: 2810
  year: 2021
  end-page: 2825
  ident: b3
  article-title: A supervised segmentation network for hyperspectral image classification
  publication-title: IEEE Trans. Image Process.
– reference: C. Herrmann, R.S. Bowen, N. Wadhwa, R. Garg, Q. He, J.T. Barron, R. Zabih, Learning to autofocus, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR, p. 2020.
– volume: 27
  start-page: 1126
  year: 2017
  end-page: 1137
  ident: b11
  article-title: Edge-based defocus blur estimation with adaptive scale selection
  publication-title: IEEE Trans. Image Process.
– volume: 56
  start-page: 27
  issue: 1
  year: 2024
  ident: 10.1016/j.neucom.2024.128880_b6
  article-title: A deep convolutional encoder–decoder–restorer architecture for image deblurring
  publication-title: Neural Process. Lett.
  doi: 10.1007/s11063-024-11455-w
– volume: 30
  start-page: 2810
  year: 2021
  ident: 10.1016/j.neucom.2024.128880_b3
  article-title: A supervised segmentation network for hyperspectral image classification
  publication-title: IEEE Trans. Image Process.
  doi: 10.1109/TIP.2021.3055613
– ident: 10.1016/j.neucom.2024.128880_b40
  doi: 10.1109/CVPR52729.2023.00570
– volume: vol. 2
  start-page: 168
  year: 1994
  ident: 10.1016/j.neucom.2024.128880_b46
  article-title: Two deterministic half-quadratic regularization algorithms for computed imaging
– start-page: 1
  year: 2020
  ident: 10.1016/j.neucom.2024.128880_b42
  article-title: Modeling defocus-disparity in dual-pixel sensors
– ident: 10.1016/j.neucom.2024.128880_b13
  doi: 10.1109/CVPR.2017.295
– ident: 10.1016/j.neucom.2024.128880_b50
– ident: 10.1016/j.neucom.2024.128880_b30
  doi: 10.1109/CVPR52688.2022.00564
– year: 2016
  ident: 10.1016/j.neucom.2024.128880_b32
  article-title: Image restoration using very deep convolutional encoder–decoder networks with symmetric skip connections
– ident: 10.1016/j.neucom.2024.128880_b28
  doi: 10.1109/ICCV48922.2021.00229
– ident: 10.1016/j.neucom.2024.128880_b33
  doi: 10.1109/CVPR46437.2021.01458
– volume: 29
  start-page: 7389
  year: 2020
  ident: 10.1016/j.neucom.2024.128880_b2
  article-title: Foveabox: Beyound anchor-based object detection
  publication-title: IEEE Trans. Image Process.
  doi: 10.1109/TIP.2020.3002345
– ident: 10.1016/j.neucom.2024.128880_b52
  doi: 10.1109/CVPR52729.2023.00557
– ident: 10.1016/j.neucom.2024.128880_b53
  doi: 10.1109/ICCV51070.2023.01158
– ident: 10.1016/j.neucom.2024.128880_b8
  doi: 10.1109/CVPR.2015.7298665
– ident: 10.1016/j.neucom.2024.128880_b55
  doi: 10.1109/CVPR.2018.00068
– ident: 10.1016/j.neucom.2024.128880_b12
  doi: 10.1109/CVPR.2019.01250
– volume: 30
  start-page: 8170
  year: 2021
  ident: 10.1016/j.neucom.2024.128880_b36
  article-title: Multi-task learning framework for motion estimation and dynamic scene deblurring
  publication-title: IEEE Trans. Image Process.
  doi: 10.1109/TIP.2021.3113185
– volume: 25
  start-page: 1660
  issue: 4
  year: 2016
  ident: 10.1016/j.neucom.2024.128880_b10
  article-title: Non-parametric blur map regression for depth of field extension
  publication-title: IEEE Trans. Image Process.
  doi: 10.1109/TIP.2016.2526907
– ident: 10.1016/j.neucom.2024.128880_b18
  doi: 10.1109/CVPR42600.2020.00230
– volume: 16
  start-page: 1440
  year: 2020
  ident: 10.1016/j.neucom.2024.128880_b23
  article-title: Single-shot face anti-spoofing for dual pixel camera
  publication-title: IEEE Trans. Inf. Forensics Secur.
  doi: 10.1109/TIFS.2020.3035879
– ident: 10.1016/j.neucom.2024.128880_b43
  doi: 10.1109/CVPR.2017.179
– ident: 10.1016/j.neucom.2024.128880_b20
  doi: 10.1109/CVPR.2019.00165
– volume: 25
  start-page: 1626
  issue: 4
  year: 2016
  ident: 10.1016/j.neucom.2024.128880_b9
  article-title: Lbp-based segmentation of defocus blur
  publication-title: IEEE Trans. Image Process.
  doi: 10.1109/TIP.2016.2528042
– ident: 10.1016/j.neucom.2024.128880_b49
  doi: 10.1109/ICCV.2015.123
– ident: 10.1016/j.neucom.2024.128880_b19
  doi: 10.1007/978-3-030-01267-0_32
– volume: 30
  start-page: 1812
  year: 2021
  ident: 10.1016/j.neucom.2024.128880_b1
  article-title: Exploring the effects of blur and deblurring to visual object tracking
  publication-title: IEEE Trans. Image Process.
  doi: 10.1109/TIP.2020.3045630
– volume: 13
  start-page: 600
  issue: 4
  year: 2004
  ident: 10.1016/j.neucom.2024.128880_b54
  article-title: Image quality assessment: from error visibility to structural similarity
  publication-title: IEEE Trans. Image Process.
  doi: 10.1109/TIP.2003.819861
– volume: 9
  start-page: 878
  issue: 5
  year: 2022
  ident: 10.1016/j.neucom.2024.128880_b29
  article-title: Bambnet: A blur-aware multi-branch network for dual-pixel defocus deblurring
  publication-title: IEEE/CAA J. Autom. Sin.
  doi: 10.1109/JAS.2022.105563
– ident: 10.1016/j.neucom.2024.128880_b48
  doi: 10.1109/CVPR.2018.00931
– start-page: 582
  year: 2020
  ident: 10.1016/j.neucom.2024.128880_b22
  article-title: Du 2 net: Learning depth estimation from dual-cameras and dual-pixels
– ident: 10.1016/j.neucom.2024.128880_b44
  doi: 10.1109/ICCV.2015.316
– volume: 1
  start-page: 11
  issue: 06
  year: 2013
  ident: 10.1016/j.neucom.2024.128880_b17
  article-title: A simple model for on-sensor phase-detection autofocusing algorithm
  publication-title: J. Comput. Commun.
  doi: 10.4236/jcc.2013.16003
– volume: 32
  start-page: 8026
  year: 2019
  ident: 10.1016/j.neucom.2024.128880_b51
  article-title: Pytorch: An imperative style, high-performance deep learning library
  publication-title: Adv. Neural Inf. Process. Syst.
– volume: 45
  start-page: 4734
  issue: 17
  year: 2020
  ident: 10.1016/j.neucom.2024.128880_b24
  article-title: Image-free real-time 3-d tracking of a fast-moving object using dual-pixel detection
  publication-title: Opt. Lett.
  doi: 10.1364/OL.399204
– volume: 28
  start-page: 2107
  issue: 5
  year: 2018
  ident: 10.1016/j.neucom.2024.128880_b37
  article-title: A local metric for defocus blur detection based on cnn feature learning
  publication-title: IEEE Trans. Image Process.
  doi: 10.1109/TIP.2018.2881830
– ident: 10.1016/j.neucom.2024.128880_b27
  doi: 10.1109/CVPR46437.2021.00432
– volume: 30
  start-page: 3419
  year: 2021
  ident: 10.1016/j.neucom.2024.128880_b38
  article-title: Joint depth and defocus estimation from a single image using physical consistency
  publication-title: IEEE Trans. Image Process.
  doi: 10.1109/TIP.2021.3061901
– ident: 10.1016/j.neucom.2024.128880_b45
  doi: 10.1109/CVPRW.2019.00247
– volume: 53
  start-page: 4419
  year: 2021
  ident: 10.1016/j.neucom.2024.128880_b7
  article-title: Stack-based scale-recurrent network for face image deblurring
  publication-title: Neural Process. Lett.
  doi: 10.1007/s11063-021-10604-9
– ident: 10.1016/j.neucom.2024.128880_b21
  doi: 10.1109/ICCV.2019.00772
– volume: 22
  start-page: 1033
  year: 2009
  ident: 10.1016/j.neucom.2024.128880_b14
  article-title: Fast image deconvolution using hyper-laplacian priors
  publication-title: Adv. Neural Inf. Process. Syst.
– year: 2024
  ident: 10.1016/j.neucom.2024.128880_b41
– volume: 27
  start-page: 1126
  issue: 3
  year: 2017
  ident: 10.1016/j.neucom.2024.128880_b11
  article-title: Edge-based defocus blur estimation with adaptive scale selection
  publication-title: IEEE Trans. Image Process.
  doi: 10.1109/TIP.2017.2771563
– volume: 15
  start-page: 5747
  issue: 3
  year: 2015
  ident: 10.1016/j.neucom.2024.128880_b16
  article-title: Sensor-based auto-focusing system using multi-scale feature extraction and phase correlation matching
  publication-title: Sensors
  doi: 10.3390/s150305747
– start-page: 111
  year: 2020
  ident: 10.1016/j.neucom.2024.128880_b26
  article-title: Defocus deblurring using dual-pixel data
– volume: 12
  start-page: 58
  issue: 1
  year: 1995
  ident: 10.1016/j.neucom.2024.128880_b15
  article-title: Blind deconvolution by means of the richardson–lucy algorithm
  publication-title: J. Opt. Soc. Amer. A
  doi: 10.1364/JOSAA.12.000058
– start-page: 234
  year: 2015
  ident: 10.1016/j.neucom.2024.128880_b31
  article-title: U-Net: convolutional networks for biomedical image segmentation
– volume: 111
  start-page: 257
  issue: 3
  year: 2023
  ident: 10.1016/j.neucom.2024.128880_b5
  article-title: Object detection in 20 years: A survey
  publication-title: Proc. IEEE
  doi: 10.1109/JPROC.2023.3238524
– ident: 10.1016/j.neucom.2024.128880_b25
  doi: 10.1109/CVPR.2019.01125
– volume: 15
  start-page: 3952
  issue: 7
  year: 2018
  ident: 10.1016/j.neucom.2024.128880_b4
  article-title: Multimodal face-pose estimation with multitask manifold deep learning
  publication-title: IEEE Trans. Ind. Inf.
  doi: 10.1109/TII.2018.2884211
– ident: 10.1016/j.neucom.2024.128880_b34
  doi: 10.1109/CVPR.2019.00613
– ident: 10.1016/j.neucom.2024.128880_b35
  doi: 10.1109/ICCV.2019.00897
– ident: 10.1016/j.neucom.2024.128880_b47
  doi: 10.1109/CVPR.2019.00953
– volume: 25
  start-page: 5943
  issue: 12
  year: 2016
  ident: 10.1016/j.neucom.2024.128880_b39
  article-title: Defocus map estimation from a single image based on two-parameter defocus model
  publication-title: IEEE Trans. Image Process.
  doi: 10.1109/TIP.2016.2617460
SSID ssj0017129
Score 2.4457517
Snippet It is a challenging task to recover sharp image from a single defocus blurry image in real-world applications. On many modern cameras, dual-pixel (DP) sensors...
SourceID crossref
elsevier
SourceType Index Database
Publisher
StartPage 128880
SubjectTerms Defocus deblurring
Image deblurring
Title Learning dual-pixel alignment for defocus deblurring
URI https://dx.doi.org/10.1016/j.neucom.2024.128880
Volume 616
WOSCitedRecordID wos001363744800001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
journalDatabaseRights – providerCode: PRVESC
  databaseName: Elsevier SD Freedom Collection Journals 2021
  issn: 0925-2312
  databaseCode: AIEXJ
  dateStart: 19950101
  customDbUrl:
  isFulltext: true
  dateEnd: 99991231
  titleUrlDefault: https://www.sciencedirect.com
  omitProxy: false
  ssIdentifier: ssj0017129
  providerName: Elsevier
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV1LT9wwELa2wKGXFloqoLTKoVejjfOwfVwhqlKtUIVA2p4ixw-U1Ta7gg1d_j3j2E6Wh1A59JKHlYwjf8nM58l4BqFvFFgB0YrjXGcMp0YozERKMaeGZMlwqKVQbbEJenbGJhP-azAYh7UwtzNa12y14ov_CjW0Adh26ewr4O6EQgMcA-iwBdhh-0_Aj4Ozw66ywotqpW0QcnXlfvvbqEKlzVw2N7AvZ9YD6I3XNCRyasCotcUevBth9MdmU1D21encBuM2CuB302kNdy5mQVqb9rGxjZOqvuuU_7lXc_P66q-u1n0OJAthyr3zENqAGT7Qo3m8rgnB7jFXo-mJknb-gulRrRsbsQMdpEf95Q9zYj-yVV0EYQhOmxZOSmGlFE7KG7RJaMZBx22OTk8mP7u_SjQmLveif_qwlLKN93v6NM9TlTX6cbGN3vl5QzRyeO-gga4_oPehJkfkVfRHlAb4ox7-qIM_AvgjD3_Uw7-LLr-fXBz_wL4yBpYwBVxiFad0qEUmM6pzlielYsCzYiINIcykimipTcJVHiuTw1eXGJIDXSl5JhJiRJJ8Qhv1vNZ7KBIq1twYQ1VZptwoAYRXMspYKcscyNw-wmEQioVLgFK8NPj7iIaRKjyJc-SsAPhfvPPglT19Rm_7d_MQbSyvG_0FbcnbZXVz_dVjfw8jWGkY
linkProvider Elsevier
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Learning+dual-pixel+alignment+for+defocus+deblurring&rft.jtitle=Neurocomputing+%28Amsterdam%29&rft.au=Li%2C+Yu&rft.au=Yi%2C+Yaling&rft.au=Shu%2C+Xinya&rft.au=Ren%2C+Dongwei&rft.date=2025-02-01&rft.issn=0925-2312&rft.volume=616&rft.spage=128880&rft_id=info:doi/10.1016%2Fj.neucom.2024.128880&rft.externalDBID=n%2Fa&rft.externalDocID=10_1016_j_neucom_2024_128880
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0925-2312&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0925-2312&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0925-2312&client=summon