Graph matching with a dual-step EM algorithm
This paper describes a new approach to matching geometric structure in 2D point-sets. The novel feature is to unify the tasks of estimating transformation geometry and identifying point-correspondence matches. Unification is realized by constructing a mixture model over the bipartite graph represent...
Saved in:
| Published in | IEEE transactions on pattern analysis and machine intelligence Vol. 20; no. 11; pp. 1236 - 1253 |
|---|---|
| Main Authors | , |
| Format | Journal Article |
| Language | English |
| Published |
IEEE
01.11.1998
|
| Subjects | |
| Online Access | Get full text |
Cover
Loading…
| Abstract | This paper describes a new approach to matching geometric structure in 2D point-sets. The novel feature is to unify the tasks of estimating transformation geometry and identifying point-correspondence matches. Unification is realized by constructing a mixture model over the bipartite graph representing the correspondence match and by affecting optimization using the EM algorithm. According to our EM framework, the probabilities of structural correspondence gate contributions to the expected likelihood function used to estimate maximum likelihood transformation parameters. These gating probabilities measure the consistency of the matched neighborhoods in the graphs. The recovery of transformational geometry and hard correspondence matches are interleaved and are realized by applying coupled update operations to the expected log-likelihood function. We evaluate the technique on two real-world problems. |
|---|---|
| AbstractList | This paper describes a new approach to matching geometric structure in 2D point-sets. The novel feature is to unify the tasks of estimating transformation geometry and identifying point-correspondence matches. Unification is realized by constructing a mixture model over the bipartite graph representing the correspondence match and by affecting optimization using the EM algorithm. According to our EM framework, the probabilities of structural correspondence gate contributions to the expected likelihood function used to estimate maximum likelihood transformation parameters. These gating probabilities measure the consistency of the matched neighborhoods in the graphs. The recovery of transformational geometry and hard correspondence matches are interleaved and are realized by applying coupled update operations to the expected log-likelihood function. We evaluate the technique on two real-world problems This paper describes a new approach to matching geometric structure in 2D point-sets. The novel feature is to unify the tasks of estimating transformation geometry and identifying point-correspondence matches. Unification is realized by constructing a mixture model over the bipartite graph representing the correspondence match and by affecting optimization using the EM algorithm. According to our EM framework, the probabilities of structural correspondence gate contributions to the expected likelihood function used to estimate maximum likelihood transformation parameters. These gating probabilities measure the consistency of the matched neighborhoods in the graphs. The recovery of transformational geometry and hard correspondence matches are interleaved and are realized by applying coupled update operations to the expected log-likelihood function. We evaluate the technique on two real-world problems. This paper describes a new approach to matching geometric structure in 2D point-sets. The novel feature is to unify the tasks of estimating transformation geometry and identifying point-correspondence matches. Unification is realized by constructing a mixture model over the bipartite graph representing the correspondence match and by affecting optimization using the EM algorithm. According to our EM framework, the probabilities of structural correspondence gate contributions to the expected likelihood function used to estimate maximum likelihood transformation parameters. These gating probabilities measure the consistency of the matched neighborhoods in the graphs. The recovery of transformational geometry and hard correspondence matches are interleaved and are realized by applying coupled update operations to the expected log-likelihood function. In this way, the two processes bootstrap one another. This provides a means of rejecting structural outliers. We evaluate the technique on two real-world problems. The fist involves the matching of different perspective views of 3.5-inch floppy discs. The second example is furnished by the matching of a digital map against aerial images that are subject to severe barrel distortion due to a line-scan sampling process. We complement these experiments with a sensitivity study based on synthetic data. |
| Author | Cross, A.D.J. Hancock, E.R. |
| Author_xml | – sequence: 1 givenname: A.D.J. surname: Cross fullname: Cross, A.D.J. organization: Dept. of Comput. Sci., York Univ., UK – sequence: 2 givenname: E.R. surname: Hancock fullname: Hancock, E.R. |
| BookMark | eNqFkM1LAzEQxXOoYFs9ePW0J0Fw28nnZo9SahUqXvQcstlsG9kvkxTxv3fLFg8ieJph5vcejzdDk7ZrLUJXGBYYQ76kbJFR4DyboClgQVIpiTxHsxDeATDjQKfobuN1v08aHc3etbvk08V9opPyoOs0RNsn6-dE17vOD_fmAp1Vug728jTn6O1h_bp6TLcvm6fV_TY1FGRMtSQMBEjAkFUUF9ZUtuQls4WW-bBjTkujpS2IyIBrwoshjSgJMwRnOVA6Rzejb--7j4MNUTUuGFvXurXdISgiKaY5yP_BjDMhxBG8HUHjuxC8rVTvXaP9l8KgjmUpytRY1sAuf7HGRR1d10avXf2n4npUOGvtj_Pp-Q1283Vd |
| CODEN | ITPIDJ |
| CitedBy_id | crossref_primary_10_1016_j_cviu_2013_09_003 crossref_primary_10_1109_91_917123 crossref_primary_10_1109_TIP_2015_2469153 crossref_primary_10_1016_j_patcog_2016_11_004 crossref_primary_10_1109_TMI_2010_2090354 crossref_primary_10_1016_j_patrec_2004_10_020 crossref_primary_10_1016_j_patrec_2007_01_020 crossref_primary_10_1007_s11771_013_1831_1 crossref_primary_10_1016_S0262_8856_02_00040_9 crossref_primary_10_1109_TPAMI_2002_1017624 crossref_primary_10_1016_j_patcog_2014_06_011 crossref_primary_10_1109_TIP_2002_801120 crossref_primary_10_1017_aer_2016_16 crossref_primary_10_1016_j_patcog_2006_08_001 crossref_primary_10_1016_j_patrec_2012_04_008 crossref_primary_10_1109_TPAMI_2010_46 crossref_primary_10_1016_S0031_3203_01_00173_X crossref_primary_10_1109_TIP_2003_816010 crossref_primary_10_1109_TIP_2005_863114 crossref_primary_10_1109_34_977570 crossref_primary_10_1109_JPHOT_2018_2876689 crossref_primary_10_1080_19479831003802790 crossref_primary_10_1016_j_imavis_2006_08_005 crossref_primary_10_1109_ACCESS_2022_3156662 crossref_primary_10_1016_S1077_3142_03_00009_2 crossref_primary_10_1016_S0031_3203_01_00232_1 crossref_primary_10_1016_j_patrec_2016_07_016 crossref_primary_10_1016_S1571_0653_04_00493_7 crossref_primary_10_1016_j_cviu_2010_07_007 crossref_primary_10_1016_j_patcog_2005_05_007 crossref_primary_10_1016_S0031_3203_00_00067_4 crossref_primary_10_1016_j_patcog_2007_06_002 crossref_primary_10_1109_TPAMI_2014_2324568 crossref_primary_10_1142_S0218001418550145 crossref_primary_10_1016_S0031_3203_02_00054_7 crossref_primary_10_1007_s11263_014_0766_9 crossref_primary_10_1118_1_2836422 crossref_primary_10_1109_TPAMI_2010_169 crossref_primary_10_1109_TPAMI_2012_99 crossref_primary_10_1016_j_patcog_2010_10_009 crossref_primary_10_1109_TPAMI_2012_51 crossref_primary_10_1198_jcgs_2009_07048 crossref_primary_10_1088_0031_9155_58_12_4315 crossref_primary_10_1109_34_954602 crossref_primary_10_1111_j_1541_0420_2010_01460_x crossref_primary_10_1016_j_ress_2017_01_008 crossref_primary_10_1142_S1793351X09000768 crossref_primary_10_12677_aam_2024_138364 crossref_primary_10_1016_j_patrec_2007_02_013 crossref_primary_10_1016_S0165_0114_01_00131_2 crossref_primary_10_1016_S0167_8655_02_00186_1 crossref_primary_10_1016_j_websem_2008_12_001 crossref_primary_10_1109_TPAMI_2003_1177149 crossref_primary_10_1109_TPAMI_2013_16 crossref_primary_10_1109_TGRS_2009_2023794 crossref_primary_10_1016_j_ins_2006_02_013 crossref_primary_10_1142_S0218001404003289 crossref_primary_10_1142_S0218213008004345 crossref_primary_10_1016_j_media_2005_08_002 crossref_primary_10_1109_4233_992159 crossref_primary_10_2139_ssrn_3199442 crossref_primary_10_1016_S0031_3203_03_00009_8 crossref_primary_10_1007_s43670_025_00097_1 crossref_primary_10_1006_cviu_1999_0746 crossref_primary_10_1016_j_sigpro_2009_04_010 crossref_primary_10_1109_TPAMI_2007_1099 crossref_primary_10_1111_j_1541_0420_2006_00622_x crossref_primary_10_1016_j_inffus_2009_12_001 crossref_primary_10_1016_j_cviu_2009_01_004 crossref_primary_10_1016_j_patrec_2011_01_015 crossref_primary_10_1109_TPAMI_2005_225 crossref_primary_10_1016_j_patrec_2009_10_011 crossref_primary_10_1109_TPAMI_2003_1251153 crossref_primary_10_1016_S1077_3142_03_00097_3 crossref_primary_10_1016_j_eswa_2014_07_051 crossref_primary_10_1109_TPAMI_2005_142 crossref_primary_10_1016_j_cviu_2011_10_009 crossref_primary_10_1016_S0262_8856_02_00010_0 |
| Cites_doi | 10.1016/0031-3203(91)90050-F 10.1109/34.329005 10.1109/CVPR.1997.609357 10.1364/JOSAA.8.000377 10.1109/34.166625 10.1006/cviu.1997.0656 10.1112/blms/16.2.81 10.1016/0262-8856(92)90043-3 10.1162/neco.1996.8.4.787 10.1109/34.601251 10.1098/rsta.1995.0022 10.7551/mitpress/3877.001.0001 10.1109/34.387502 10.1109/ICCV.1995.466883 10.1007/BFb0028337 10.1016/0004-3702(90)90098-K 10.1109/CVPR.1991.139700 10.1006/cviu.1996.0037 10.1109/34.308475 10.1162/089976698300017188 10.1109/12.210173 10.1109/34.485529 10.1109/ICCV.1995.466816 10.1145/357346.357349 10.1162/neco.1994.6.2.181 10.1023/A:1007940112931 10.1016/0734-189X(89)90146-1 10.1109/TPAMI.1982.4767255 10.1016/S0262-8856(97)00014-0 10.1109/34.546255 10.1023/A:1007927408552 10.1109/34.6778 10.1007/BF02028352 10.1016/0262-8856(95)99727-I 10.1098/rspb.1991.0045 10.1006/cviu.1995.1004 10.1016/0734-189X(85)90126-4 10.1111/j.2517-6161.1977.tb01600.x 10.1016/0031-3203(86)90029-4 10.1109/34.88573 10.1109/34.584098 10.1109/34.192485 |
| ContentType | Journal Article |
| DBID | RIA RIE AAYXX CITATION 7SC 8FD JQ2 L7M L~C L~D |
| DOI | 10.1109/34.730557 |
| DatabaseName | IEEE All-Society Periodicals Package (ASPP) 1998–Present IEEE Electronic Library (IEL) CrossRef Computer and Information Systems Abstracts Technology Research Database ProQuest Computer Science Collection Advanced Technologies Database with Aerospace Computer and Information Systems Abstracts Academic Computer and Information Systems Abstracts Professional |
| DatabaseTitle | CrossRef Computer and Information Systems Abstracts Technology Research Database Computer and Information Systems Abstracts – Academic Advanced Technologies Database with Aerospace ProQuest Computer Science Collection Computer and Information Systems Abstracts Professional |
| DatabaseTitleList | Computer and Information Systems Abstracts Computer and Information Systems Abstracts |
| Database_xml | – sequence: 1 dbid: RIE name: IEEE Electronic Library (IEL) url: https://proxy.k.utb.cz/login?url=https://ieeexplore.ieee.org/ sourceTypes: Publisher |
| DeliveryMethod | fulltext_linktorsrc |
| Discipline | Engineering Computer Science |
| EndPage | 1253 |
| ExternalDocumentID | 10_1109_34_730557 730557 |
| GroupedDBID | --- -DZ -~X .DC 0R~ 29I 4.4 53G 5GY 5VS 6IK 97E 9M8 AAJGR AARMG AASAJ AAWTH ABAZT ABFSI ABQJQ ABVLG ACGFO ACGFS ACIWK ACNCT ADRHT AENEX AETEA 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 FA8 HZ~ H~9 IBMZZ ICLAB IEDLZ IFIPE IFJZH IPLJI JAVBF LAI M43 MS~ O9- OCL P2P PQQKQ RIA RIE RNI RNS RXW RZB TAE TN5 UHB VH1 XJT ~02 AAYOK AAYXX CITATION RIG 7SC 8FD JQ2 L7M L~C L~D |
| ID | FETCH-LOGICAL-c308t-a82406080107f31becfed5d4eba89cfe153dca8eb26705a25b0146d24c2179033 |
| IEDL.DBID | RIE |
| ISSN | 0162-8828 |
| IngestDate | Thu Jul 10 17:44:46 EDT 2025 Fri Jul 11 01:10:04 EDT 2025 Thu Apr 24 22:51:57 EDT 2025 Tue Jul 01 03:18:13 EDT 2025 Wed Aug 27 02:47:52 EDT 2025 |
| IsPeerReviewed | true |
| IsScholarly | true |
| Issue | 11 |
| Language | English |
| License | https://ieeexplore.ieee.org/Xplorehelp/downloads/license-information/IEEE.html |
| LinkModel | DirectLink |
| MergedId | FETCHMERGED-LOGICAL-c308t-a82406080107f31becfed5d4eba89cfe153dca8eb26705a25b0146d24c2179033 |
| Notes | ObjectType-Article-2 SourceType-Scholarly Journals-1 ObjectType-Feature-1 content type line 23 |
| PQID | 27546668 |
| PQPubID | 23500 |
| PageCount | 18 |
| ParticipantIDs | crossref_primary_10_1109_34_730557 proquest_miscellaneous_28313908 ieee_primary_730557 proquest_miscellaneous_27546668 crossref_citationtrail_10_1109_34_730557 |
| ProviderPackageCode | CITATION AAYXX |
| PublicationCentury | 1900 |
| PublicationDate | 1998-11-01 |
| PublicationDateYYYYMMDD | 1998-11-01 |
| PublicationDate_xml | – month: 11 year: 1998 text: 1998-11-01 day: 01 |
| PublicationDecade | 1990 |
| PublicationTitle | IEEE transactions on pattern analysis and machine intelligence |
| PublicationTitleAbbrev | TPAMI |
| PublicationYear | 1998 |
| Publisher | IEEE |
| Publisher_xml | – name: IEEE |
| References | bibi12369 bibi12368 bibi12367 ullman (bibi123639) 1979 bibi12366 bibi12365 bibi12364 bibi12363 bibi123640 bibi123641 bibi123626 bibi123627 bibi123624 bibi123625 bibi123622 bibi123644 bibi123623 bibi123645 bibi123620 bibi123642 bibi123621 cross (bibi123610) 1998; 10 ahuja (bibi12362) 1985; 29 bibi123628 bibi123629 bibi123630 utans (bibi123643) 1993 bibi123637 bibi123615 bibi123638 bibi123616 bibi123635 bibi123613 hempel (bibi123619) 1986 bibi123636 bibi123614 bibi123611 bibi123633 bibi123634 bibi123631 bibi123632 bibi12361 dempster (bibi123612) 1977; 39 bibi123617 bibi123618 |
| References_xml | – year: 1986 ident: bibi123619 publication-title: Robust Statistics The Approach Based on Influence Functions – ident: bibi123638 doi: 10.1016/0031-3203(91)90050-F – ident: bibi123617 doi: 10.1109/34.329005 – ident: bibi123632 doi: 10.1109/CVPR.1997.609357 – ident: bibi123625 doi: 10.1364/JOSAA.8.000377 – ident: bibi123611 doi: 10.1109/34.166625 – ident: bibi123644 doi: 10.1006/cviu.1997.0656 – ident: bibi123624 doi: 10.1112/blms/16.2.81 – ident: bibi123635 doi: 10.1016/0262-8856(92)90043-3 – ident: bibi123615 doi: 10.1162/neco.1996.8.4.787 – ident: bibi123645 doi: 10.1109/34.601251 – ident: bibi123636 doi: 10.1098/rsta.1995.0022 – year: 1979 ident: bibi123639 publication-title: The Interpretation of Visual Motion doi: 10.7551/mitpress/3877.001.0001 – ident: bibi123634 doi: 10.1109/34.387502 – ident: bibi123621 doi: 10.1109/ICCV.1995.466883 – ident: bibi12366 doi: 10.1007/BFb0028337 – ident: bibi123613 doi: 10.1016/0004-3702(90)90098-K – ident: bibi123622 doi: 10.1109/CVPR.1991.139700 – ident: bibi123629 doi: 10.1006/cviu.1996.0037 – ident: bibi12364 doi: 10.1109/34.308475 – ident: bibi123614 doi: 10.1162/089976698300017188 – ident: bibi123626 doi: 10.1109/12.210173 – ident: bibi12365 doi: 10.1109/34.485529 – ident: bibi123618 doi: 10.1109/ICCV.1995.466816 – ident: bibi12367 doi: 10.1145/357346.357349 – ident: bibi123623 doi: 10.1162/neco.1994.6.2.181 – ident: bibi123620 doi: 10.1023/A:1007940112931 – ident: bibi12363 doi: 10.1016/0734-189X(89)90146-1 – ident: bibi12361 doi: 10.1109/TPAMI.1982.4767255 – ident: bibi123628 doi: 10.1016/S0262-8856(97)00014-0 – ident: bibi123627 doi: 10.1109/34.546255 – ident: bibi123637 doi: 10.1023/A:1007927408552 – ident: bibi123640 doi: 10.1109/34.6778 – ident: bibi123616 doi: 10.1007/BF02028352 – ident: bibi12369 doi: 10.1016/0262-8856(95)99727-I – ident: bibi123633 doi: 10.1098/rspb.1991.0045 – year: 1993 ident: bibi123643 article-title: mixture models and the em algorithms for object recognition within compositional hierarchies – ident: bibi12368 doi: 10.1006/cviu.1995.1004 – volume: 29 start-page: 286 year: 1985 ident: bibi12362 article-title: image representation using voronoi tessellation publication-title: Computer Vision Graphics and Image Processing doi: 10.1016/0734-189X(85)90126-4 – volume: 39 start-page: 1 year: 1977 ident: bibi123612 article-title: maximum-likelihood from incomplete data via the em algorithm publication-title: J Royal Statistical Soc Series B (methodological) doi: 10.1111/j.2517-6161.1977.tb01600.x – ident: bibi123630 doi: 10.1016/0031-3203(86)90029-4 – ident: bibi123641 doi: 10.1109/34.88573 – ident: bibi123631 doi: 10.1109/34.584098 – volume: 10 start-page: 780 year: 1998 ident: bibi123610 article-title: perspective pose recovery with a dual step em algorithm publication-title: Advances in neural information processing systems – ident: bibi123642 doi: 10.1109/34.192485 |
| SSID | ssj0014503 |
| Score | 2.0206664 |
| Snippet | This paper describes a new approach to matching geometric structure in 2D point-sets. The novel feature is to unify the tasks of estimating transformation... |
| SourceID | proquest crossref ieee |
| SourceType | Aggregation Database Enrichment Source Index Database Publisher |
| StartPage | 1236 |
| SubjectTerms | Bipartite graph Computational geometry Computer vision Image sampling Information geometry Matrix decomposition Maximum likelihood estimation Pattern matching Robot sensing systems |
| Title | Graph matching with a dual-step EM algorithm |
| URI | https://ieeexplore.ieee.org/document/730557 https://www.proquest.com/docview/27546668 https://www.proquest.com/docview/28313908 |
| Volume | 20 |
| hasFullText | 1 |
| inHoldings | 1 |
| isFullTextHit | |
| isPrint | |
| link | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwjV07T8MwELZoJxgoFBDlaSEGBpImsd04I0ItFVKZqNQt8isMlKZqk4Vfz9lJKl5CbJZ1jiI_v7Pv-w6h65AaKSTlXkaU9mgUa0_QTHk8NFJzbWhALXd48jQYT-njjM1qnW3HhTHGuOAz49uie8vXuSrtVVk_tvJUcQu1wG-rqFqbBwPKXBJkADCwwMGLqEWEwiDpE-pXDb8cPS6Xyo8N2J0qo05F1147MUIbTPLql4X01fs3qcZ__vAe2q3RJb6rpsM-2jKLLuo0mRtwvZC7aOeTDOEBun2wqtUYsKsLrMT2bhYLbFlaHkyCJR5OsJi_5CuofztE09Hw-X7s1VkUPEUCXniC20MbgCE4ehkJYcwyo5m2Y8QTKMOWp5Xg4GEP4oCJiEmrJ6MjqiKr3kXIEWov8oU5RphJbRJlI2mIphmjUpkkZCJTPNIJ4IAeumk6OFW1xLjNdDFPnasRJCmhadUnPXS1MV1Wuhq_GXVtn24MmtrLZtBSWAv2gUMsTF6u0yhmFNwx_ocFJwB5A37y65dP0bajHDqq4RlqF6vSnAPmKOSFm20fhhHUEg |
| linkProvider | IEEE |
| linkToHtml | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwzV07T8MwED4hGICBRwFRnhYCiYG0iWMTZ2BAvAq0TCCxBcd2GCgtglYI_gp_hR_H2UkqHhUbEptlWZHjO5-_s---A9gMmEllyoSXhUp7jEbakyxTnghMqoU2zGc2d7h1sdu4YmfX_HoE3ga5MMYYF3xmarbp3vJ1V_XtVVk9svRUURFBeW5entE_e9o7PURhblF6fHR50PCKEgKeCn3R86SwJxaiIvRysjDACWdGc20nKGJs437XSgp0L3cjn0vKU0umoilT1FJX2dtOtO9jCDM4zZPDBk8UjLuyywiZ0KSg31LQFgV-XA9ZLZ_ql8POVW_5YfLdOXY8De_lCuThK3e1fi-tqddv5JD_dIlmYKrAz2Q_V_hZGDGdCkyXtSlIYaoqMPmJaHEOdk4sLzdBdO5CR4m9fSaS2Dw0D9X8gRy1iGzfdh-x_34erv7kDxZgtNPtmEUgPNUmVjZWKNQs4yxVJg64zJSgOkakU4XtUqCJKkjUbS2PduKcKT9OQpbkMqjCxmDoQ84cMmxQxcpwMKDsXS-VJMHdbp9wZMd0-08JjThDh1P8MkKECOp9sTT0y-sw3rhsNZPm6cX5Mky4BEuXWLkCo73HvllFhNVL15ymE7j5a735ACziMPI |
| 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=Graph+matching+with+a+dual-step+EM+algorithm&rft.jtitle=IEEE+transactions+on+pattern+analysis+and+machine+intelligence&rft.au=Cross%2C+A.D.J.&rft.au=Hancock%2C+E.R.&rft.date=1998-11-01&rft.pub=IEEE&rft.issn=0162-8828&rft.volume=20&rft.issue=11&rft.spage=1236&rft.epage=1253&rft_id=info:doi/10.1109%2F34.730557&rft.externalDocID=730557 |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0162-8828&client=summon |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0162-8828&client=summon |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0162-8828&client=summon |