Fast Domain Decomposition for Global Image Smoothing

Edge-preserving smoothing (EPS) can be formulated as minimizing an objective function that consists of data and regularization terms. At the price of high-computational cost, this global EPS approach is more robust and versatile than a local one that typically has a form of weighted averaging. In th...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:IEEE transactions on image processing Ročník 26; číslo 8; s. 4079 - 4091
Hlavní autori: Kim, Youngjung, Min, Dongbo, Ham, Bumsub, Sohn, Kwanghoon
Médium: Journal Article
Jazyk:English
Vydavateľské údaje: IEEE 01.08.2017
Predmet:
ISSN:1057-7149, 1941-0042
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Abstract Edge-preserving smoothing (EPS) can be formulated as minimizing an objective function that consists of data and regularization terms. At the price of high-computational cost, this global EPS approach is more robust and versatile than a local one that typically has a form of weighted averaging. In this paper, we introduce an efficient decomposition-based method for global EPS that minimizes the objective function of L2 data and (possibly non-smooth and non-convex) regularization terms in linear time. Different from previous decompositionbased methods, which require solving a large linear system, our approach solves an equivalent constrained optimization problem, resulting in a sequence of 1-D sub-problems. This enables applying fast linear time solver for weighted-least squares and -L1 smoothing problems. An alternating direction method of multipliers algorithm is adopted to guarantee fast convergence. Our method is fully parallelizable, and its runtime is even comparable to the state-of-the-art local EPS approaches. We also propose a family of fast majorization-minimization algorithms that minimize an objective with non-convex regularization terms. Experimental results demonstrate the effectiveness and flexibility of our approach in a range of image processing and computational photography applications.
AbstractList Edge-preserving smoothing (EPS) can be formulated as minimizing an objective function that consists of data and regularization terms. At the price of high-computational cost, this global EPS approach is more robust and versatile than a local one that typically has a form of weighted averaging. In this paper, we introduce an efficient decomposition-based method for global EPS that minimizes the objective function of L2 data and (possibly non-smooth and non-convex) regularization terms in linear time. Different from previous decompositionbased methods, which require solving a large linear system, our approach solves an equivalent constrained optimization problem, resulting in a sequence of 1-D sub-problems. This enables applying fast linear time solver for weighted-least squares and -L1 smoothing problems. An alternating direction method of multipliers algorithm is adopted to guarantee fast convergence. Our method is fully parallelizable, and its runtime is even comparable to the state-of-the-art local EPS approaches. We also propose a family of fast majorization-minimization algorithms that minimize an objective with non-convex regularization terms. Experimental results demonstrate the effectiveness and flexibility of our approach in a range of image processing and computational photography applications.
Author Dongbo Min
Kwanghoon Sohn
Bumsub Ham
Youngjung Kim
Author_xml – sequence: 1
  givenname: Youngjung
  surname: Kim
  fullname: Kim, Youngjung
– sequence: 2
  givenname: Dongbo
  orcidid: 0000-0003-4825-5240
  surname: Min
  fullname: Min, Dongbo
– sequence: 3
  givenname: Bumsub
  surname: Ham
  fullname: Ham, Bumsub
– sequence: 4
  givenname: Kwanghoon
  orcidid: 0000-0002-4288-0160
  surname: Sohn
  fullname: Sohn, Kwanghoon
BookMark eNp9z01Lw0AQgOFFKthW74KX_IHEmf3MHqW1NVBQsJ7DJtmtK0m2ZHPx35vS4sGDp5nLO8OzILM-9JaQe4QMEfTjvnjLKKDKqEKQFK_IHDXHFIDT2bSDUKlCrm_IIsYvAOQC5ZzwjYljsg6d8X2ytnXojiH60Yc-cWFItm2oTJsUnTnY5L0LYfz0_eGWXDvTRnt3mUvysXner17S3eu2WD3t0ppRPaayyZk0gDoHUeUCKXXAtJE1BUe5tpZVumkYKmS1VNI6EDkaYYTLBVh0bEngfLceQoyDdeVx8J0ZvkuE8qQuJ3V5UpcX9ZTIP0ntR3PyjIPx7X_hwzn01trfP0ozlTPOfgDnD2TB
CODEN IIPRE4
CitedBy_id crossref_primary_10_1109_TNSRE_2021_3123394
crossref_primary_10_1007_s11042_023_16914_5
crossref_primary_10_1109_TOH_2022_3207928
crossref_primary_10_1016_j_patcog_2020_107670
crossref_primary_10_1016_j_dsp_2022_103623
crossref_primary_10_1017_S1431927621000106
crossref_primary_10_1109_TGRS_2020_3004224
crossref_primary_10_4018_IJGHPC_2018070102
crossref_primary_10_1007_s00371_018_1504_z
crossref_primary_10_1109_TIP_2019_2941326
crossref_primary_10_1049_iet_ipr_2019_0484
crossref_primary_10_1016_j_ins_2019_03_056
crossref_primary_10_1016_j_patcog_2022_108900
crossref_primary_10_3390_math11102389
crossref_primary_10_1049_iet_ipr_2019_1577
crossref_primary_10_1109_TIP_2023_3247181
crossref_primary_10_1016_j_fsidi_2022_301473
crossref_primary_10_1109_TGRS_2024_3490773
crossref_primary_10_1007_s10851_019_00918_8
crossref_primary_10_1145_3388887
Cites_doi 10.1007/978-3-642-15549-9_1
10.1137/120896219
10.1145/1882261.1866171
10.1017/CBO9780511804441
10.1109/CVPR.2014.362
10.1561/0600000058
10.1109/TPAMI.2007.1177
10.1145/2010324.1964964
10.1145/1276377.1276497
10.1109/ICCV.1998.710815
10.1145/2461912.2461992
10.1145/2766946
10.1109/ICCV.2001.937655
10.1109/CVPR.2012.6247743
10.1145/2185520.2185529
10.1109/TIP.2014.2366600
10.1007/978-3-319-10602-1_3
10.1007/s11263-007-0110-8
10.1137/1.9780898718003
10.1007/978-1-84996-290-2
10.1109/TIP.2013.2253479
10.1109/TIP.2003.819861
10.1145/2070781.2024208
10.1109/TIP.2010.2047910
10.1109/TIP.2011.2163164
10.1109/ICCV.2013.13
10.1109/83.392335
10.1145/1141911.1142018
10.1007/s101070100263
10.1145/2366145.2366158
10.1145/1360612.1360666
10.1561/2200000016
10.1137/080724265
10.1109/LSP.2013.2278339
10.1007/978-3-540-88690-7_20
10.1109/ICCV.2015.32
10.1145/1778765.1778837
10.1145/1276377.1276441
10.1111/j.1467-8659.2009.01645.x
10.1137/140971518
10.1109/ICIP.2015.7350827
10.1145/1015706.1015780
10.1109/CVPR.2015.7299115
10.1023/A:1007963824710
ContentType Journal Article
DBID 97E
RIA
RIE
AAYXX
CITATION
DOI 10.1109/TIP.2017.2710621
DatabaseName IEEE Xplore (IEEE)
IEEE All-Society Periodicals Package (ASPP) 1998–Present
IEEE Electronic Library (IEL)
CrossRef
DatabaseTitle CrossRef
DatabaseTitleList
Database_xml – sequence: 1
  dbid: RIE
  name: IEEE Xplore
  url: https://ieeexplore.ieee.org/
  sourceTypes: Publisher
DeliveryMethod fulltext_linktorsrc
Discipline Applied Sciences
Engineering
EISSN 1941-0042
EndPage 4091
ExternalDocumentID 10_1109_TIP_2017_2710621
7937834
Genre orig-research
GrantInformation_xml – fundername: National Research Foundation of Korea through the MSIP
  grantid: 2017R1C1B2005584
  funderid: 10.13039/501100003725
– fundername: Institute for Information and communications Technology Promotion Grant through the Korea Government (MSIP)
  grantid: 2016-0-00197
GroupedDBID ---
-~X
.DC
0R~
29I
4.4
53G
5GY
5VS
6IK
97E
AAJGR
AARMG
AASAJ
AAWTH
ABAZT
ABFSI
ABQJQ
ABVLG
ACGFO
ACGFS
ACIWK
AENEX
AETIX
AGQYO
AGSQL
AHBIQ
AI.
AIBXA
AKJIK
AKQYR
ALLEH
ALMA_UNASSIGNED_HOLDINGS
ASUFR
ATWAV
BEFXN
BFFAM
BGNUA
BKEBE
BPEOZ
CS3
DU5
E.L
EBS
EJD
F5P
HZ~
H~9
ICLAB
IFIPE
IFJZH
IPLJI
JAVBF
LAI
M43
MS~
O9-
OCL
P2P
RIA
RIE
RNS
TAE
TN5
VH1
AAYXX
CITATION
ID FETCH-LOGICAL-c329t-6d836a019805b85122f039a6c20f249ee3b9dd31713c676ef0581a5a5f850e1f3
IEDL.DBID RIE
ISICitedReferencesCount 20
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000404286300007&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 1057-7149
IngestDate Sat Nov 29 03:21:06 EST 2025
Tue Nov 18 22:25:23 EST 2025
Tue Aug 26 17:00:45 EDT 2025
IsPeerReviewed true
IsScholarly true
Issue 8
Language English
License https://ieeexplore.ieee.org/Xplorehelp/downloads/license-information/IEEE.html
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-c329t-6d836a019805b85122f039a6c20f249ee3b9dd31713c676ef0581a5a5f850e1f3
ORCID 0000-0002-4288-0160
0000-0003-4825-5240
PageCount 13
ParticipantIDs crossref_primary_10_1109_TIP_2017_2710621
ieee_primary_7937834
crossref_citationtrail_10_1109_TIP_2017_2710621
PublicationCentury 2000
PublicationDate 2017-Aug.
2017-8-00
PublicationDateYYYYMMDD 2017-08-01
PublicationDate_xml – month: 08
  year: 2017
  text: 2017-Aug.
PublicationDecade 2010
PublicationTitle IEEE transactions on image processing
PublicationTitleAbbrev TIP
PublicationYear 2017
Publisher IEEE
Publisher_xml – name: IEEE
References ref13
ref12
(ref49) 2016
ref53
ref52
ref17
ref16
winnemöller (ref28) 2006; 25
bi (ref36) 2015; 34
ref50
liu (ref30) 2006
krishnan (ref19) 2013; 32
grady (ref9) 2010
ref46
ref45
ref47
ref41
ref44
ref43
(ref48) 2016
fattal (ref29) 2007; 26
ref7
yang (ref15) 2009
xu (ref4) 2012; 31
ref3
ref6
ref5
zhang (ref51) 2014
ref40
ref35
ref34
ref37
gastal (ref18) 2011; 30
ref31
ref33
xu (ref39) 2011; 30
ref32
ref2
ref1
ref38
hong (ref24) 2012
kass (ref10) 2010; 29
kopf (ref14) 2007; 26
nesterov (ref27) 1983; 27
ref23
ref26
ref25
ref20
schwarz (ref42) 1988
ref22
ref21
min (ref11) 2012; 21
he (ref8) 2009
References_xml – ident: ref17
  doi: 10.1007/978-3-642-15549-9_1
– ident: ref23
  doi: 10.1137/120896219
– ident: ref35
  doi: 10.1145/1882261.1866171
– ident: ref53
  doi: 10.1017/CBO9780511804441
– ident: ref33
  doi: 10.1109/CVPR.2014.362
– start-page: 1
  year: 2014
  ident: ref51
  article-title: Rolling guidance filter
  publication-title: Proc Eur Conf Comput Vis
– ident: ref44
  doi: 10.1561/0600000058
– ident: ref7
  doi: 10.1109/TPAMI.2007.1177
– volume: 30
  start-page: 1
  year: 2011
  ident: ref18
  article-title: Domain transform for edge-aware image and video processing
  publication-title: ACM Trans Graph
  doi: 10.1145/2010324.1964964
– volume: 26
  start-page: 96
  year: 2007
  ident: ref14
  article-title: Joint bilateral upsampling
  publication-title: ACM Trans Graph
  doi: 10.1145/1276377.1276497
– year: 2016
  ident: ref48
  publication-title: SuiteSparse
– ident: ref12
  doi: 10.1109/ICCV.1998.710815
– volume: 32
  start-page: 142-1
  year: 2013
  ident: ref19
  article-title: Efficient preconditioning of Laplacian matrices for computer graphics
  publication-title: ACM Trans Graph
  doi: 10.1145/2461912.2461992
– volume: 34
  start-page: 778
  year: 2015
  ident: ref36
  article-title: An $L_{1}$ image transform for edge-preserving smoothing and scene-level intrinsic decomposition
  publication-title: ACM Trans Graph
  doi: 10.1145/2766946
– ident: ref50
  doi: 10.1109/ICCV.2001.937655
– ident: ref6
  doi: 10.1109/CVPR.2012.6247743
– ident: ref32
  doi: 10.1145/2185520.2185529
– ident: ref22
  doi: 10.1109/TIP.2014.2366600
– ident: ref25
  doi: 10.1007/978-3-319-10602-1_3
– ident: ref16
  doi: 10.1007/s11263-007-0110-8
– ident: ref40
  doi: 10.1137/1.9780898718003
– year: 2010
  ident: ref9
  publication-title: Discrete Calculus Applied Analysis on Graphs for Computational Science
  doi: 10.1007/978-1-84996-290-2
– ident: ref3
  doi: 10.1109/TIP.2013.2253479
– ident: ref45
  doi: 10.1109/TIP.2003.819861
– volume: 30
  start-page: 174
  year: 2011
  ident: ref39
  article-title: Image smoothing via $L_{0}$ gradient minimization
  publication-title: ACM Trans Graph
  doi: 10.1145/2070781.2024208
– ident: ref21
  doi: 10.1109/TIP.2010.2047910
– volume: 21
  start-page: 1176
  year: 2012
  ident: ref11
  article-title: Depth video enhancement based on weighted mode filtering
  publication-title: IEEE Trans Image Process
  doi: 10.1109/TIP.2011.2163164
– ident: ref2
  doi: 10.1109/ICCV.2013.13
– ident: ref41
  doi: 10.1109/83.392335
– volume: 25
  start-page: 1221
  year: 2006
  ident: ref28
  article-title: Real-time video abstraction
  publication-title: ACM Trans Graph
  doi: 10.1145/1141911.1142018
– ident: ref47
  doi: 10.1007/s101070100263
– volume: 31
  start-page: 139
  year: 2012
  ident: ref4
  article-title: Structure extraction from texture via relative total variation
  publication-title: ACM Trans Graph
  doi: 10.1145/2366145.2366158
– ident: ref1
  doi: 10.1145/1360612.1360666
– start-page: 557
  year: 2009
  ident: ref15
  article-title: Real-time O(1) bilateral filtering
  publication-title: Proc IEEE Conf Comput Vis Pattern Recognit
– start-page: 1956
  year: 2009
  ident: ref8
  article-title: Single image haze removal using dark channel prior
  publication-title: Proc IEEE Conf Comput Vis Pattern Recognit
– year: 1988
  ident: ref42
  publication-title: Numerische Mathematik
– start-page: 901
  year: 2006
  ident: ref30
  article-title: Noise estimation from a single image
  publication-title: Proc IEEE Conf Comput Vis Pattern Recognit
– ident: ref26
  doi: 10.1561/2200000016
– ident: ref20
  doi: 10.1137/080724265
– ident: ref43
  doi: 10.1109/LSP.2013.2278339
– ident: ref5
  doi: 10.1007/978-3-540-88690-7_20
– ident: ref52
  doi: 10.1109/ICCV.2015.32
– year: 2016
  ident: ref49
  publication-title: FFTW
– volume: 29
  start-page: 100
  year: 2010
  ident: ref10
  article-title: Smoothed local histogram filters
  publication-title: ACM Trans Graph
  doi: 10.1145/1778765.1778837
– volume: 26
  start-page: 51
  year: 2007
  ident: ref29
  article-title: Multiscale shape and detail enhancement from multi-light image collections
  publication-title: ACM Trans Graph
  doi: 10.1145/1276377.1276441
– ident: ref31
  doi: 10.1111/j.1467-8659.2009.01645.x
– ident: ref46
  doi: 10.1137/140971518
– ident: ref38
  doi: 10.1109/ICIP.2015.7350827
– ident: ref34
  doi: 10.1145/1015706.1015780
– volume: 27
  start-page: 372
  year: 1983
  ident: ref27
  article-title: A method of solving a convex programming problem with convergence rate O( $1/k^{2}$ )
  publication-title: Soviet Math Dokl
– ident: ref37
  doi: 10.1109/CVPR.2015.7299115
– ident: ref13
  doi: 10.1023/A:1007963824710
– year: 2012
  ident: ref24
  publication-title: On the linear convergence of the alternating direction method of multipliers
SSID ssj0014516
Score 2.3720062
Snippet Edge-preserving smoothing (EPS) can be formulated as minimizing an objective function that consists of data and regularization terms. At the price of...
SourceID crossref
ieee
SourceType Enrichment Source
Index Database
Publisher
StartPage 4079
SubjectTerms Acceleration
alternating minimization
Edge-preserving image smoothing
Histograms
joint image filtering
Linear programming
Linear systems
majorization-minimization algorithm
Optimization
Photography
Smoothing methods
weighted-least squares
Title Fast Domain Decomposition for Global Image Smoothing
URI https://ieeexplore.ieee.org/document/7937834
Volume 26
WOSCitedRecordID wos000404286300007&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: PRVIEE
  databaseName: IEEE Xplore
  customDbUrl:
  eissn: 1941-0042
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0014516
  issn: 1057-7149
  databaseCode: RIE
  dateStart: 19920101
  isFulltext: true
  titleUrlDefault: https://ieeexplore.ieee.org/
  providerName: IEEE
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV3PS8MwFH7M4UEPTjfF-YscvAh2S9omTY_iHO4yBk7YraT5AQPXytb595u0XZkggrdS8qB8Sfu-9L18H8C9ZMqYNLXMDSvqhVIZL1a-8gQTqWULOCSSl2YT0XTKF4t41oLH5iyM1rpsPtMDd1nW8lUut-5X2dBpufEgPICDKGLVWa2mYuAMZ8vKJo28yNL-XUkSx8P5ZOZ6uKKBb9Mp88mPFLTnqVKmlHHnfw9zCic1dURP1VyfQUtnXejUNBLVL-mmC8d7GoM9CMdiU6BRvhLLDI20ayGv-7SQ5auo0vxHk5X9rqC3VW4nzoadw_v4Zf786tVWCZ4M_LjwmOIBE5aucUxTS6J83-AgFkz62NgNltZBGitluQIJJIuYNphyIqighlOsiQkuoJ3lmb4EhI2RlndxZlN9GGqeKqklCRRlShDC0j4Md-glstYRd3YWH0m5n8BxYvFOHN5JjXcfHpqIz0pD44-xPQd1M65G-er329dw5IKrhrwbaBfrrb6FQ_lVLDfru3KBfAMB-7b9
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1bS8MwGP2YU1AfnG6K85oHXwS7pZek6aM4x4ZzDJywt5LmAgPXyS7-fpO2KxNE8K2UJJSTNt9Jv5PzAdwJKrVOEsPcsCROIKR2IulJh1OeGLaAA1ewrNhEOByyySQaVeChPAujlMrEZ6plL7NcvpyLtf1V1rZebswPdmCXBIGH89NaZc7AlpzNcpskdEJD_DdJSRy1x_2RVXGFLc8EVOq5P4LQVlWVLKh0a_97nGM4Ksgjesxn-wQqKq1DrSCSqPhMl3U43HIZbEDQ5csV6sxnfJqijrIi8kKphQxjRbnrP-rPzMqC3mZzM3Wm2ym8d5_HTz2nKJbgCN-LVg6VzKfcEDaGSWJolOdp7EecCg9rs8VSyk8iKQ1bcH1BQ6o0JszlhBPNCFau9s-gms5TdQ4Iay0M82LUBPsgUCyRQgnXl4RK7ro0aUJ7g14sCidxW9DiI852FDiKDd6xxTsu8G7CfdnjM3fR-KNtw0JdtitQvvj99i3s98avg3jQH75cwoEdKJfnXUF1tVira9gTX6vpcnGTvSzft9O6RA
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=Fast+Domain+Decomposition+for+Global+Image+Smoothing&rft.jtitle=IEEE+transactions+on+image+processing&rft.au=Kim%2C+Youngjung&rft.au=Min%2C+Dongbo&rft.au=Ham%2C+Bumsub&rft.au=Sohn%2C+Kwanghoon&rft.date=2017-08-01&rft.issn=1057-7149&rft.eissn=1941-0042&rft.volume=26&rft.issue=8&rft.spage=4079&rft.epage=4091&rft_id=info:doi/10.1109%2FTIP.2017.2710621&rft.externalDBID=n%2Fa&rft.externalDocID=10_1109_TIP_2017_2710621
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1057-7149&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1057-7149&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1057-7149&client=summon