Learning CRFs Using Graph Cuts View Full Text


Ontology type: schema:Chapter      Open Access: True


Chapter Info

DATE

2008

AUTHORS

Martin Szummer , Pushmeet Kohli , Derek Hoiem

ABSTRACT

Many computer vision problems are naturally formulated as random fields, specifically MRFs or CRFs. The introduction of graph cuts has enabled efficient and optimal inference in associative random fields, greatly advancing applications such as segmentation, stereo reconstruction and many others. However, while fast inference is now widespread, parameter learning in random fields has remained an intractable problem. This paper shows how to apply fast inference algorithms, in particular graph cuts, to learn parameters of random fields with similar efficiency. We find optimal parameter values under standard regularized objective functions that ensure good generalization. Our algorithm enables learning of many parameters in reasonable time, and we explore further speedup techniques. We also discuss extensions to non-associative and multi-class problems. We evaluate the method on image segmentation and geometry recognition. More... »

PAGES

582-595

References to SciGraph publications

Book

TITLE

Computer Vision – ECCV 2008

ISBN

978-3-540-88685-3
978-3-540-88688-4

Identifiers

URI

http://scigraph.springernature.com/pub.10.1007/978-3-540-88688-4_43

DOI

http://dx.doi.org/10.1007/978-3-540-88688-4_43

DIMENSIONS

https://app.dimensions.ai/details/publication/pub.1052807962


Indexing Status Check whether this publication has been indexed by Scopus and Web Of Science using the SN Indexing Status Tool
Incoming Citations Browse incoming citations for this publication using opencitations.net

JSON-LD is the canonical representation for SciGraph data.

TIP: You can open this SciGraph record using an external JSON-LD service: JSON-LD Playground Google SDTT

[
  {
    "@context": "https://springernature.github.io/scigraph/jsonld/sgcontext.json", 
    "about": [
      {
        "id": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/0801", 
        "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
        "name": "Artificial Intelligence and Image Processing", 
        "type": "DefinedTerm"
      }, 
      {
        "id": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/08", 
        "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
        "name": "Information and Computing Sciences", 
        "type": "DefinedTerm"
      }
    ], 
    "author": [
      {
        "affiliation": {
          "alternateName": "Microsoft Research (United Kingdom)", 
          "id": "https://www.grid.ac/institutes/grid.24488.32", 
          "name": [
            "Microsoft Research, CB3 0FB, Cambridge, United Kingdom"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Szummer", 
        "givenName": "Martin", 
        "id": "sg:person.015100145701.84", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.015100145701.84"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Microsoft Research (United Kingdom)", 
          "id": "https://www.grid.ac/institutes/grid.24488.32", 
          "name": [
            "Microsoft Research, CB3 0FB, Cambridge, United Kingdom"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Kohli", 
        "givenName": "Pushmeet", 
        "id": "sg:person.0606005504.21", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.0606005504.21"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "University of Illinois at Urbana Champaign", 
          "id": "https://www.grid.ac/institutes/grid.35403.31", 
          "name": [
            "Beckman Institute, University of Illinois at Urbana-Champaign, USA"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Hoiem", 
        "givenName": "Derek", 
        "id": "sg:person.01302474045.37", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01302474045.37"
        ], 
        "type": "Person"
      }
    ], 
    "citation": [
      {
        "id": "https://doi.org/10.1145/1390156.1390195", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1018948353"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1007/11744047_2", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1021485887", 
          "https://doi.org/10.1007/11744047_2"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1007/11744047_2", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1021485887", 
          "https://doi.org/10.1007/11744047_2"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1145/1102351.1102464", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1022920538"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1007/11585978_11", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1039535601", 
          "https://doi.org/10.1007/11585978_11"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1007/11585978_11", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1039535601", 
          "https://doi.org/10.1007/11585978_11"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1016/s0166-218x(01)00341-9", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1039547005"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1145/1015330.1015444", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1040268511"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1007/11744047_3", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1040981648", 
          "https://doi.org/10.1007/11744047_3"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1007/11744047_3", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1040981648", 
          "https://doi.org/10.1007/11744047_3"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1109/34.969114", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1061157335"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1109/tpami.2004.1262177", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1061742646"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1109/tpami.2005.161", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1061742822"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1145/1015706.1015720", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1063148832"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1109/cvpr.2007.383095", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1093678308"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1109/iccv.2005.107", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1094730474"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1109/iccv.2001.937505", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1095383001"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1109/cvpr.2005.133", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1095487088"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.3115/1118693.1118694", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1099201575"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.3115/1118693.1118694", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1099201575"
        ], 
        "type": "CreativeWork"
      }
    ], 
    "datePublished": "2008", 
    "datePublishedReg": "2008-01-01", 
    "description": "Many computer vision problems are naturally formulated as random fields, specifically MRFs or CRFs. The introduction of graph cuts has enabled efficient and optimal inference in associative random fields, greatly advancing applications such as segmentation, stereo reconstruction and many others. However, while fast inference is now widespread, parameter learning in random fields has remained an intractable problem. This paper shows how to apply fast inference algorithms, in particular graph cuts, to learn parameters of random fields with similar efficiency. We find optimal parameter values under standard regularized objective functions that ensure good generalization. Our algorithm enables learning of many parameters in reasonable time, and we explore further speedup techniques. We also discuss extensions to non-associative and multi-class problems. We evaluate the method on image segmentation and geometry recognition.", 
    "editor": [
      {
        "familyName": "Forsyth", 
        "givenName": "David", 
        "type": "Person"
      }, 
      {
        "familyName": "Torr", 
        "givenName": "Philip", 
        "type": "Person"
      }, 
      {
        "familyName": "Zisserman", 
        "givenName": "Andrew", 
        "type": "Person"
      }
    ], 
    "genre": "chapter", 
    "id": "sg:pub.10.1007/978-3-540-88688-4_43", 
    "inLanguage": [
      "en"
    ], 
    "isAccessibleForFree": true, 
    "isPartOf": {
      "isbn": [
        "978-3-540-88685-3", 
        "978-3-540-88688-4"
      ], 
      "name": "Computer Vision \u2013 ECCV 2008", 
      "type": "Book"
    }, 
    "name": "Learning CRFs Using Graph Cuts", 
    "pagination": "582-595", 
    "productId": [
      {
        "name": "doi", 
        "type": "PropertyValue", 
        "value": [
          "10.1007/978-3-540-88688-4_43"
        ]
      }, 
      {
        "name": "readcube_id", 
        "type": "PropertyValue", 
        "value": [
          "fb8a520c71ff339aff85776edf9911d053ef54615d6416d364564f171b055506"
        ]
      }, 
      {
        "name": "dimensions_id", 
        "type": "PropertyValue", 
        "value": [
          "pub.1052807962"
        ]
      }
    ], 
    "publisher": {
      "location": "Berlin, Heidelberg", 
      "name": "Springer Berlin Heidelberg", 
      "type": "Organisation"
    }, 
    "sameAs": [
      "https://doi.org/10.1007/978-3-540-88688-4_43", 
      "https://app.dimensions.ai/details/publication/pub.1052807962"
    ], 
    "sdDataset": "chapters", 
    "sdDatePublished": "2019-04-16T06:10", 
    "sdLicense": "https://scigraph.springernature.com/explorer/license/", 
    "sdPublisher": {
      "name": "Springer Nature - SN SciGraph project", 
      "type": "Organization"
    }, 
    "sdSource": "s3://com-uberresearch-data-dimensions-target-20181106-alternative/cleanup/v134/2549eaecd7973599484d7c17b260dba0a4ecb94b/merge/v9/a6c9fde33151104705d4d7ff012ea9563521a3ce/jats-lookup/v90/0000000350_0000000350/records_77564_00000001.jsonl", 
    "type": "Chapter", 
    "url": "https://link.springer.com/10.1007%2F978-3-540-88688-4_43"
  }
]
 

Download the RDF metadata as:  json-ld nt turtle xml License info

HOW TO GET THIS DATA PROGRAMMATICALLY:

JSON-LD is a popular format for linked data which is fully compatible with JSON.

curl -H 'Accept: application/ld+json' 'https://scigraph.springernature.com/pub.10.1007/978-3-540-88688-4_43'

N-Triples is a line-based linked data format ideal for batch operations.

curl -H 'Accept: application/n-triples' 'https://scigraph.springernature.com/pub.10.1007/978-3-540-88688-4_43'

Turtle is a human-readable linked data format.

curl -H 'Accept: text/turtle' 'https://scigraph.springernature.com/pub.10.1007/978-3-540-88688-4_43'

RDF/XML is a standard XML format for linked data.

curl -H 'Accept: application/rdf+xml' 'https://scigraph.springernature.com/pub.10.1007/978-3-540-88688-4_43'


 

This table displays all metadata directly associated to this object as RDF triples.

143 TRIPLES      23 PREDICATES      43 URIs      20 LITERALS      8 BLANK NODES

Subject Predicate Object
1 sg:pub.10.1007/978-3-540-88688-4_43 schema:about anzsrc-for:08
2 anzsrc-for:0801
3 schema:author N22f3750a247040e9b59cf303b8b29330
4 schema:citation sg:pub.10.1007/11585978_11
5 sg:pub.10.1007/11744047_2
6 sg:pub.10.1007/11744047_3
7 https://doi.org/10.1016/s0166-218x(01)00341-9
8 https://doi.org/10.1109/34.969114
9 https://doi.org/10.1109/cvpr.2005.133
10 https://doi.org/10.1109/cvpr.2007.383095
11 https://doi.org/10.1109/iccv.2001.937505
12 https://doi.org/10.1109/iccv.2005.107
13 https://doi.org/10.1109/tpami.2004.1262177
14 https://doi.org/10.1109/tpami.2005.161
15 https://doi.org/10.1145/1015330.1015444
16 https://doi.org/10.1145/1015706.1015720
17 https://doi.org/10.1145/1102351.1102464
18 https://doi.org/10.1145/1390156.1390195
19 https://doi.org/10.3115/1118693.1118694
20 schema:datePublished 2008
21 schema:datePublishedReg 2008-01-01
22 schema:description Many computer vision problems are naturally formulated as random fields, specifically MRFs or CRFs. The introduction of graph cuts has enabled efficient and optimal inference in associative random fields, greatly advancing applications such as segmentation, stereo reconstruction and many others. However, while fast inference is now widespread, parameter learning in random fields has remained an intractable problem. This paper shows how to apply fast inference algorithms, in particular graph cuts, to learn parameters of random fields with similar efficiency. We find optimal parameter values under standard regularized objective functions that ensure good generalization. Our algorithm enables learning of many parameters in reasonable time, and we explore further speedup techniques. We also discuss extensions to non-associative and multi-class problems. We evaluate the method on image segmentation and geometry recognition.
23 schema:editor N1fdef2afb1fe45b9bb25ce807af77528
24 schema:genre chapter
25 schema:inLanguage en
26 schema:isAccessibleForFree true
27 schema:isPartOf N0ec9011c5a6046329a043af0dafdcb9c
28 schema:name Learning CRFs Using Graph Cuts
29 schema:pagination 582-595
30 schema:productId N25d30479e88846429653af5472cd83cf
31 N483dcc48dc30434ba4944425e3745f71
32 N63e6ac53f63b4935be439deecde2bdd8
33 schema:publisher N97627163df2d4ad78872c88e222a06e9
34 schema:sameAs https://app.dimensions.ai/details/publication/pub.1052807962
35 https://doi.org/10.1007/978-3-540-88688-4_43
36 schema:sdDatePublished 2019-04-16T06:10
37 schema:sdLicense https://scigraph.springernature.com/explorer/license/
38 schema:sdPublisher Nc9be0b64abeb4715a329c3fd963ab146
39 schema:url https://link.springer.com/10.1007%2F978-3-540-88688-4_43
40 sgo:license sg:explorer/license/
41 sgo:sdDataset chapters
42 rdf:type schema:Chapter
43 N0ec9011c5a6046329a043af0dafdcb9c schema:isbn 978-3-540-88685-3
44 978-3-540-88688-4
45 schema:name Computer Vision – ECCV 2008
46 rdf:type schema:Book
47 N1fdef2afb1fe45b9bb25ce807af77528 rdf:first N6e7e639f64684dcab1692dddc4baefbb
48 rdf:rest N8dcd34287f7f4b8792905dcf3a23dc5f
49 N22f3750a247040e9b59cf303b8b29330 rdf:first sg:person.015100145701.84
50 rdf:rest N25704e02f9bf45e58b81b468b6ea4233
51 N25704e02f9bf45e58b81b468b6ea4233 rdf:first sg:person.0606005504.21
52 rdf:rest N504045bfd0db41bfac134841da1000d2
53 N25d30479e88846429653af5472cd83cf schema:name doi
54 schema:value 10.1007/978-3-540-88688-4_43
55 rdf:type schema:PropertyValue
56 N483dcc48dc30434ba4944425e3745f71 schema:name dimensions_id
57 schema:value pub.1052807962
58 rdf:type schema:PropertyValue
59 N504045bfd0db41bfac134841da1000d2 rdf:first sg:person.01302474045.37
60 rdf:rest rdf:nil
61 N63e6ac53f63b4935be439deecde2bdd8 schema:name readcube_id
62 schema:value fb8a520c71ff339aff85776edf9911d053ef54615d6416d364564f171b055506
63 rdf:type schema:PropertyValue
64 N6e7e639f64684dcab1692dddc4baefbb schema:familyName Forsyth
65 schema:givenName David
66 rdf:type schema:Person
67 N75208d6fac294d36818b4f15f0f9016f rdf:first Ndcd140286ad940f3bc5b0c445c4f91c9
68 rdf:rest rdf:nil
69 N8dcd34287f7f4b8792905dcf3a23dc5f rdf:first Na2ed2b7f46b04e6a82dc845f2c223d58
70 rdf:rest N75208d6fac294d36818b4f15f0f9016f
71 N97627163df2d4ad78872c88e222a06e9 schema:location Berlin, Heidelberg
72 schema:name Springer Berlin Heidelberg
73 rdf:type schema:Organisation
74 Na2ed2b7f46b04e6a82dc845f2c223d58 schema:familyName Torr
75 schema:givenName Philip
76 rdf:type schema:Person
77 Nc9be0b64abeb4715a329c3fd963ab146 schema:name Springer Nature - SN SciGraph project
78 rdf:type schema:Organization
79 Ndcd140286ad940f3bc5b0c445c4f91c9 schema:familyName Zisserman
80 schema:givenName Andrew
81 rdf:type schema:Person
82 anzsrc-for:08 schema:inDefinedTermSet anzsrc-for:
83 schema:name Information and Computing Sciences
84 rdf:type schema:DefinedTerm
85 anzsrc-for:0801 schema:inDefinedTermSet anzsrc-for:
86 schema:name Artificial Intelligence and Image Processing
87 rdf:type schema:DefinedTerm
88 sg:person.01302474045.37 schema:affiliation https://www.grid.ac/institutes/grid.35403.31
89 schema:familyName Hoiem
90 schema:givenName Derek
91 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01302474045.37
92 rdf:type schema:Person
93 sg:person.015100145701.84 schema:affiliation https://www.grid.ac/institutes/grid.24488.32
94 schema:familyName Szummer
95 schema:givenName Martin
96 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.015100145701.84
97 rdf:type schema:Person
98 sg:person.0606005504.21 schema:affiliation https://www.grid.ac/institutes/grid.24488.32
99 schema:familyName Kohli
100 schema:givenName Pushmeet
101 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.0606005504.21
102 rdf:type schema:Person
103 sg:pub.10.1007/11585978_11 schema:sameAs https://app.dimensions.ai/details/publication/pub.1039535601
104 https://doi.org/10.1007/11585978_11
105 rdf:type schema:CreativeWork
106 sg:pub.10.1007/11744047_2 schema:sameAs https://app.dimensions.ai/details/publication/pub.1021485887
107 https://doi.org/10.1007/11744047_2
108 rdf:type schema:CreativeWork
109 sg:pub.10.1007/11744047_3 schema:sameAs https://app.dimensions.ai/details/publication/pub.1040981648
110 https://doi.org/10.1007/11744047_3
111 rdf:type schema:CreativeWork
112 https://doi.org/10.1016/s0166-218x(01)00341-9 schema:sameAs https://app.dimensions.ai/details/publication/pub.1039547005
113 rdf:type schema:CreativeWork
114 https://doi.org/10.1109/34.969114 schema:sameAs https://app.dimensions.ai/details/publication/pub.1061157335
115 rdf:type schema:CreativeWork
116 https://doi.org/10.1109/cvpr.2005.133 schema:sameAs https://app.dimensions.ai/details/publication/pub.1095487088
117 rdf:type schema:CreativeWork
118 https://doi.org/10.1109/cvpr.2007.383095 schema:sameAs https://app.dimensions.ai/details/publication/pub.1093678308
119 rdf:type schema:CreativeWork
120 https://doi.org/10.1109/iccv.2001.937505 schema:sameAs https://app.dimensions.ai/details/publication/pub.1095383001
121 rdf:type schema:CreativeWork
122 https://doi.org/10.1109/iccv.2005.107 schema:sameAs https://app.dimensions.ai/details/publication/pub.1094730474
123 rdf:type schema:CreativeWork
124 https://doi.org/10.1109/tpami.2004.1262177 schema:sameAs https://app.dimensions.ai/details/publication/pub.1061742646
125 rdf:type schema:CreativeWork
126 https://doi.org/10.1109/tpami.2005.161 schema:sameAs https://app.dimensions.ai/details/publication/pub.1061742822
127 rdf:type schema:CreativeWork
128 https://doi.org/10.1145/1015330.1015444 schema:sameAs https://app.dimensions.ai/details/publication/pub.1040268511
129 rdf:type schema:CreativeWork
130 https://doi.org/10.1145/1015706.1015720 schema:sameAs https://app.dimensions.ai/details/publication/pub.1063148832
131 rdf:type schema:CreativeWork
132 https://doi.org/10.1145/1102351.1102464 schema:sameAs https://app.dimensions.ai/details/publication/pub.1022920538
133 rdf:type schema:CreativeWork
134 https://doi.org/10.1145/1390156.1390195 schema:sameAs https://app.dimensions.ai/details/publication/pub.1018948353
135 rdf:type schema:CreativeWork
136 https://doi.org/10.3115/1118693.1118694 schema:sameAs https://app.dimensions.ai/details/publication/pub.1099201575
137 rdf:type schema:CreativeWork
138 https://www.grid.ac/institutes/grid.24488.32 schema:alternateName Microsoft Research (United Kingdom)
139 schema:name Microsoft Research, CB3 0FB, Cambridge, United Kingdom
140 rdf:type schema:Organization
141 https://www.grid.ac/institutes/grid.35403.31 schema:alternateName University of Illinois at Urbana Champaign
142 schema:name Beckman Institute, University of Illinois at Urbana-Champaign, USA
143 rdf:type schema:Organization
 




Preview window. Press ESC to close (or click here)


...