Hierarchical Support Vector Random Fields: Joint Training to Combine Local and Global Features View Full Text


Ontology type: schema:Chapter     


Chapter Info

DATE

2008

AUTHORS

Paul Schnitzspan , Mario Fritz , Bernt Schiele

ABSTRACT

Recently, impressive results have been reported for the detection of objects in challenging real-world scenes. Interestingly however, the underlying models vary greatly even between the most successful approaches. Methods using a global feature descriptor (e.g. ) paired with discriminative classifiers such as SVMs enable high levels of performance, but require large amounts of training data and typically degrade in the presence of partial occlusions. Local feature-based approaches (e.g. ) are more robust in the presence of partial occlusions but often produce a significant number of false positives. This paper proposes a novel approach called hierarchical support vector random field that allows 1) to combine the power of global feature-based approaches with the flexibility of local feature-based methods in one consistent multi-layer framework and 2) to automatically learn the tradeoff and the optimal interplay between local, semi-local and global feature contributions. Experiments show that both the combination of local and global features as well as the joint training result in improved detection performance on challenging datasets. More... »

PAGES

527-540

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_39

DOI

http://dx.doi.org/10.1007/978-3-540-88688-4_39

DIMENSIONS

https://app.dimensions.ai/details/publication/pub.1024185122


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/2008/08", 
        "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
        "name": "Information and Computing Sciences", 
        "type": "DefinedTerm"
      }, 
      {
        "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"
      }
    ], 
    "author": [
      {
        "affiliation": {
          "alternateName": "Computer Science Department, TU Darmstadt, Germany", 
          "id": "http://www.grid.ac/institutes/grid.6546.1", 
          "name": [
            "Computer Science Department, TU Darmstadt, Germany"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Schnitzspan", 
        "givenName": "Paul", 
        "id": "sg:person.014557762621.00", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.014557762621.00"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Computer Science Department, TU Darmstadt, Germany", 
          "id": "http://www.grid.ac/institutes/grid.6546.1", 
          "name": [
            "Computer Science Department, TU Darmstadt, Germany"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Fritz", 
        "givenName": "Mario", 
        "id": "sg:person.013361072755.17", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.013361072755.17"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Computer Science Department, TU Darmstadt, Germany", 
          "id": "http://www.grid.ac/institutes/grid.6546.1", 
          "name": [
            "Computer Science Department, TU Darmstadt, Germany"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Schiele", 
        "givenName": "Bernt", 
        "id": "sg:person.01174260421.90", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01174260421.90"
        ], 
        "type": "Person"
      }
    ], 
    "datePublished": "2008", 
    "datePublishedReg": "2008-01-01", 
    "description": "Recently, impressive results have been reported for the detection of objects in challenging real-world scenes. Interestingly however, the underlying models vary greatly even between the most successful approaches. Methods using a global feature descriptor (e.g. ) paired with discriminative classifiers such as SVMs enable high levels of performance, but require large amounts of training data and typically degrade in the presence of partial occlusions. Local feature-based approaches (e.g. ) are more robust in the presence of partial occlusions but often produce a significant number of false positives. This paper proposes a novel approach called hierarchical support vector random field that allows 1) to combine the power of global feature-based approaches with the flexibility of local feature-based methods in one consistent multi-layer framework and 2) to automatically learn the tradeoff and the optimal interplay between local, semi-local and global feature contributions. Experiments show that both the combination of local and global features as well as the joint training result in improved detection performance on challenging datasets.", 
    "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_39", 
    "isAccessibleForFree": false, 
    "isPartOf": {
      "isbn": [
        "978-3-540-88685-3", 
        "978-3-540-88688-4"
      ], 
      "name": "Computer Vision \u2013 ECCV 2008", 
      "type": "Book"
    }, 
    "keywords": [
      "feature-based approach", 
      "local feature-based approach", 
      "global features", 
      "improved detection performance", 
      "partial occlusion", 
      "local feature-based methods", 
      "global feature descriptor", 
      "multi-layer framework", 
      "real-world scenes", 
      "feature-based methods", 
      "detection of objects", 
      "feature descriptors", 
      "training data", 
      "detection performance", 
      "discriminative classifier", 
      "feature contributions", 
      "joint training", 
      "impressive results", 
      "performance", 
      "optimal interplay", 
      "training results", 
      "random fields", 
      "novel approach", 
      "false positives", 
      "large amount", 
      "successful approach", 
      "method", 
      "power", 
      "classifier", 
      "SVM", 
      "approach", 
      "dataset", 
      "results", 
      "scene", 
      "flexibility", 
      "field", 
      "tradeoff", 
      "descriptors", 
      "features", 
      "objects", 
      "experiments", 
      "framework", 
      "model", 
      "amount", 
      "vector random fields", 
      "positives", 
      "detection", 
      "combination", 
      "presence", 
      "training", 
      "occlusion", 
      "data", 
      "contribution", 
      "number", 
      "significant number", 
      "interplay", 
      "high levels", 
      "levels", 
      "locals", 
      "paper"
    ], 
    "name": "Hierarchical Support Vector Random Fields: Joint Training to Combine Local and Global Features", 
    "pagination": "527-540", 
    "productId": [
      {
        "name": "dimensions_id", 
        "type": "PropertyValue", 
        "value": [
          "pub.1024185122"
        ]
      }, 
      {
        "name": "doi", 
        "type": "PropertyValue", 
        "value": [
          "10.1007/978-3-540-88688-4_39"
        ]
      }
    ], 
    "publisher": {
      "name": "Springer Nature", 
      "type": "Organisation"
    }, 
    "sameAs": [
      "https://doi.org/10.1007/978-3-540-88688-4_39", 
      "https://app.dimensions.ai/details/publication/pub.1024185122"
    ], 
    "sdDataset": "chapters", 
    "sdDatePublished": "2022-11-24T21:14", 
    "sdLicense": "https://scigraph.springernature.com/explorer/license/", 
    "sdPublisher": {
      "name": "Springer Nature - SN SciGraph project", 
      "type": "Organization"
    }, 
    "sdSource": "s3://com-springernature-scigraph/baseset/20221124/entities/gbq_results/chapter/chapter_234.jsonl", 
    "type": "Chapter", 
    "url": "https://doi.org/10.1007/978-3-540-88688-4_39"
  }
]
 

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_39'

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_39'

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_39'

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_39'


 

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

143 TRIPLES      22 PREDICATES      85 URIs      78 LITERALS      7 BLANK NODES

Subject Predicate Object
1 sg:pub.10.1007/978-3-540-88688-4_39 schema:about anzsrc-for:08
2 anzsrc-for:0801
3 schema:author N94ead2e729aa420fae9f034a4507e481
4 schema:datePublished 2008
5 schema:datePublishedReg 2008-01-01
6 schema:description Recently, impressive results have been reported for the detection of objects in challenging real-world scenes. Interestingly however, the underlying models vary greatly even between the most successful approaches. Methods using a global feature descriptor (e.g. ) paired with discriminative classifiers such as SVMs enable high levels of performance, but require large amounts of training data and typically degrade in the presence of partial occlusions. Local feature-based approaches (e.g. ) are more robust in the presence of partial occlusions but often produce a significant number of false positives. This paper proposes a novel approach called hierarchical support vector random field that allows 1) to combine the power of global feature-based approaches with the flexibility of local feature-based methods in one consistent multi-layer framework and 2) to automatically learn the tradeoff and the optimal interplay between local, semi-local and global feature contributions. Experiments show that both the combination of local and global features as well as the joint training result in improved detection performance on challenging datasets.
7 schema:editor Nf3a79c5f7a414696b7e55a4d4ea3225e
8 schema:genre chapter
9 schema:isAccessibleForFree false
10 schema:isPartOf N00bdb76800d34ca29b443e9da2d9528a
11 schema:keywords SVM
12 amount
13 approach
14 classifier
15 combination
16 contribution
17 data
18 dataset
19 descriptors
20 detection
21 detection of objects
22 detection performance
23 discriminative classifier
24 experiments
25 false positives
26 feature contributions
27 feature descriptors
28 feature-based approach
29 feature-based methods
30 features
31 field
32 flexibility
33 framework
34 global feature descriptor
35 global features
36 high levels
37 impressive results
38 improved detection performance
39 interplay
40 joint training
41 large amount
42 levels
43 local feature-based approach
44 local feature-based methods
45 locals
46 method
47 model
48 multi-layer framework
49 novel approach
50 number
51 objects
52 occlusion
53 optimal interplay
54 paper
55 partial occlusion
56 performance
57 positives
58 power
59 presence
60 random fields
61 real-world scenes
62 results
63 scene
64 significant number
65 successful approach
66 tradeoff
67 training
68 training data
69 training results
70 vector random fields
71 schema:name Hierarchical Support Vector Random Fields: Joint Training to Combine Local and Global Features
72 schema:pagination 527-540
73 schema:productId N3d58a5957bd84ff69abe338abf1828a7
74 Nf9e5d9d670a14cbaa0b1a87abb50b35a
75 schema:publisher N444a4d328fed4461a448e5aa802fab37
76 schema:sameAs https://app.dimensions.ai/details/publication/pub.1024185122
77 https://doi.org/10.1007/978-3-540-88688-4_39
78 schema:sdDatePublished 2022-11-24T21:14
79 schema:sdLicense https://scigraph.springernature.com/explorer/license/
80 schema:sdPublisher Ne2be4fbbaf144522ab6b097ae9dd7c39
81 schema:url https://doi.org/10.1007/978-3-540-88688-4_39
82 sgo:license sg:explorer/license/
83 sgo:sdDataset chapters
84 rdf:type schema:Chapter
85 N00bdb76800d34ca29b443e9da2d9528a schema:isbn 978-3-540-88685-3
86 978-3-540-88688-4
87 schema:name Computer Vision – ECCV 2008
88 rdf:type schema:Book
89 N3d58a5957bd84ff69abe338abf1828a7 schema:name doi
90 schema:value 10.1007/978-3-540-88688-4_39
91 rdf:type schema:PropertyValue
92 N444a4d328fed4461a448e5aa802fab37 schema:name Springer Nature
93 rdf:type schema:Organisation
94 N675bb46995194163974b493b8430ed9b schema:familyName Torr
95 schema:givenName Philip
96 rdf:type schema:Person
97 N872570cc0c244aab968f80b83d57fdbb rdf:first sg:person.013361072755.17
98 rdf:rest Nd9fbecce55d5467eaa28594c4ca13078
99 N94ead2e729aa420fae9f034a4507e481 rdf:first sg:person.014557762621.00
100 rdf:rest N872570cc0c244aab968f80b83d57fdbb
101 Nbff3cdbe70d04a0093332ca665da82b3 rdf:first N675bb46995194163974b493b8430ed9b
102 rdf:rest Ne62131f729424424a134a6d7b014289d
103 Nc5f2b1f44c09456fadd7b91fd33e2b7e schema:familyName Zisserman
104 schema:givenName Andrew
105 rdf:type schema:Person
106 Nc6fa9f2fd0fe4b9693e9994ec567eb12 schema:familyName Forsyth
107 schema:givenName David
108 rdf:type schema:Person
109 Nd9fbecce55d5467eaa28594c4ca13078 rdf:first sg:person.01174260421.90
110 rdf:rest rdf:nil
111 Ne2be4fbbaf144522ab6b097ae9dd7c39 schema:name Springer Nature - SN SciGraph project
112 rdf:type schema:Organization
113 Ne62131f729424424a134a6d7b014289d rdf:first Nc5f2b1f44c09456fadd7b91fd33e2b7e
114 rdf:rest rdf:nil
115 Nf3a79c5f7a414696b7e55a4d4ea3225e rdf:first Nc6fa9f2fd0fe4b9693e9994ec567eb12
116 rdf:rest Nbff3cdbe70d04a0093332ca665da82b3
117 Nf9e5d9d670a14cbaa0b1a87abb50b35a schema:name dimensions_id
118 schema:value pub.1024185122
119 rdf:type schema:PropertyValue
120 anzsrc-for:08 schema:inDefinedTermSet anzsrc-for:
121 schema:name Information and Computing Sciences
122 rdf:type schema:DefinedTerm
123 anzsrc-for:0801 schema:inDefinedTermSet anzsrc-for:
124 schema:name Artificial Intelligence and Image Processing
125 rdf:type schema:DefinedTerm
126 sg:person.01174260421.90 schema:affiliation grid-institutes:grid.6546.1
127 schema:familyName Schiele
128 schema:givenName Bernt
129 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01174260421.90
130 rdf:type schema:Person
131 sg:person.013361072755.17 schema:affiliation grid-institutes:grid.6546.1
132 schema:familyName Fritz
133 schema:givenName Mario
134 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.013361072755.17
135 rdf:type schema:Person
136 sg:person.014557762621.00 schema:affiliation grid-institutes:grid.6546.1
137 schema:familyName Schnitzspan
138 schema:givenName Paul
139 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.014557762621.00
140 rdf:type schema:Person
141 grid-institutes:grid.6546.1 schema:alternateName Computer Science Department, TU Darmstadt, Germany
142 schema:name Computer Science Department, TU Darmstadt, Germany
143 rdf:type schema:Organization
 




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


...