Active Metric Learning for Object Recognition View Full Text


Ontology type: schema:Chapter     


Chapter Info

DATE

2012

AUTHORS

Sandra Ebert , Mario Fritz , Bernt Schiele

ABSTRACT

Popular visual representations like SIFT have shown broad applicability across many task. This great generality comes naturally with a lack of specificity when focusing on a particular task or a set of classes. Metric learning approaches have been proposed to tailor general purpose representations to the needs of more specific tasks and have shown strong improvements on visual matching and recognition benchmarks. However, the performance of metric learning depends strongly on the labels that are used for learning. Therefore, we propose to combine metric learning with an active sample selection strategy in order to find labels that are representative for each class as well as improve the class separation of the learnt metric. We analyze several active sample selection strategies in terms of exploration and exploitation trade-offs. Our novel scheme achieves on three different datasets up to 10% improvement of the learned metric. We compare a batch version of our scheme to an interleaved execution of sample selection and metric learning which leads to an overall improvement of up to 23% on challenging datasets for object class recognition. More... »

PAGES

327-336

Book

TITLE

Pattern Recognition

ISBN

978-3-642-32716-2
978-3-642-32717-9

Identifiers

URI

http://scigraph.springernature.com/pub.10.1007/978-3-642-32717-9_33

DOI

http://dx.doi.org/10.1007/978-3-642-32717-9_33

DIMENSIONS

https://app.dimensions.ai/details/publication/pub.1026458093


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/2008/17", 
        "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
        "name": "Psychology and Cognitive Sciences", 
        "type": "DefinedTerm"
      }, 
      {
        "id": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/1701", 
        "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
        "name": "Psychology", 
        "type": "DefinedTerm"
      }
    ], 
    "author": [
      {
        "affiliation": {
          "alternateName": "Max Planck Institute for Informatics, Saarbrucken, Germany", 
          "id": "http://www.grid.ac/institutes/grid.419528.3", 
          "name": [
            "Max Planck Institute for Informatics, Saarbrucken, Germany"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Ebert", 
        "givenName": "Sandra", 
        "id": "sg:person.011333635343.49", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.011333635343.49"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Max Planck Institute for Informatics, Saarbrucken, Germany", 
          "id": "http://www.grid.ac/institutes/grid.419528.3", 
          "name": [
            "Max Planck Institute for Informatics, Saarbrucken, 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": "Max Planck Institute for Informatics, Saarbrucken, Germany", 
          "id": "http://www.grid.ac/institutes/grid.419528.3", 
          "name": [
            "Max Planck Institute for Informatics, Saarbrucken, 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": "2012", 
    "datePublishedReg": "2012-01-01", 
    "description": "Popular visual representations like SIFT have shown broad applicability across many task. This great generality comes naturally with a lack of specificity when focusing on a particular task or a set of classes. Metric learning approaches have been proposed to tailor general purpose representations to the needs of more specific tasks and have shown strong improvements on visual matching and recognition benchmarks. However, the performance of metric learning depends strongly on the labels that are used for learning. Therefore, we propose to combine metric learning with an active sample selection strategy in order to find labels that are representative for each class as well as improve the class separation of the learnt metric. We analyze several active sample selection strategies in terms of exploration and exploitation trade-offs. Our novel scheme achieves on three different datasets up to 10% improvement of the learned metric. We compare a batch version of our scheme to an interleaved execution of sample selection and metric learning which leads to an overall improvement of up to 23% on challenging datasets for object class recognition.", 
    "editor": [
      {
        "familyName": "Pinz", 
        "givenName": "Axel", 
        "type": "Person"
      }, 
      {
        "familyName": "Pock", 
        "givenName": "Thomas", 
        "type": "Person"
      }, 
      {
        "familyName": "Bischof", 
        "givenName": "Horst", 
        "type": "Person"
      }, 
      {
        "familyName": "Leberl", 
        "givenName": "Franz", 
        "type": "Person"
      }
    ], 
    "genre": "chapter", 
    "id": "sg:pub.10.1007/978-3-642-32717-9_33", 
    "isAccessibleForFree": false, 
    "isPartOf": {
      "isbn": [
        "978-3-642-32716-2", 
        "978-3-642-32717-9"
      ], 
      "name": "Pattern Recognition", 
      "type": "Book"
    }, 
    "keywords": [
      "object class recognition", 
      "general-purpose representation", 
      "visual matching", 
      "popular visual representation", 
      "object recognition", 
      "sample selection strategy", 
      "metric learning", 
      "recognition benchmarks", 
      "particular task", 
      "visual representation", 
      "task", 
      "specific tasks", 
      "learning", 
      "class recognition", 
      "metric learning approach", 
      "learning approach", 
      "recognition", 
      "strong improvement", 
      "representation", 
      "terms of exploration", 
      "different datasets", 
      "labels", 
      "overall improvement", 
      "sample selection", 
      "batch version", 
      "lack of specificity", 
      "strategies", 
      "version", 
      "generality", 
      "improvement", 
      "dataset", 
      "performance", 
      "execution", 
      "exploration", 
      "greater generality", 
      "need", 
      "set of classes", 
      "selection strategy", 
      "metrics", 
      "matching", 
      "lack", 
      "approach", 
      "terms", 
      "set", 
      "interleaved execution", 
      "class", 
      "benchmarks", 
      "selection", 
      "broad applicability", 
      "order", 
      "specificity", 
      "SIFT", 
      "class separation", 
      "applicability", 
      "exploitation", 
      "separation", 
      "novel scheme", 
      "scheme"
    ], 
    "name": "Active Metric Learning for Object Recognition", 
    "pagination": "327-336", 
    "productId": [
      {
        "name": "dimensions_id", 
        "type": "PropertyValue", 
        "value": [
          "pub.1026458093"
        ]
      }, 
      {
        "name": "doi", 
        "type": "PropertyValue", 
        "value": [
          "10.1007/978-3-642-32717-9_33"
        ]
      }
    ], 
    "publisher": {
      "name": "Springer Nature", 
      "type": "Organisation"
    }, 
    "sameAs": [
      "https://doi.org/10.1007/978-3-642-32717-9_33", 
      "https://app.dimensions.ai/details/publication/pub.1026458093"
    ], 
    "sdDataset": "chapters", 
    "sdDatePublished": "2022-12-01T06:50", 
    "sdLicense": "https://scigraph.springernature.com/explorer/license/", 
    "sdPublisher": {
      "name": "Springer Nature - SN SciGraph project", 
      "type": "Organization"
    }, 
    "sdSource": "s3://com-springernature-scigraph/baseset/20221201/entities/gbq_results/chapter/chapter_288.jsonl", 
    "type": "Chapter", 
    "url": "https://doi.org/10.1007/978-3-642-32717-9_33"
  }
]
 

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.10.1007/978-3-642-32717-9_33'

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

curl -H 'Accept: application/n-triples' 'https://scigraph.springernature.com/pub.10.1007/978-3-642-32717-9_33'

Turtle is a human-readable linked data format.

curl -H 'Accept: text/turtle' 'https://scigraph.springernature.com/pub.10.1007/978-3-642-32717-9_33'

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

curl -H 'Accept: application/rdf+xml' 'https://scigraph.springernature.com/pub.10.1007/978-3-642-32717-9_33'


 

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

146 TRIPLES      22 PREDICATES      83 URIs      76 LITERALS      7 BLANK NODES

Subject Predicate Object
1 sg:pub.10.1007/978-3-642-32717-9_33 schema:about anzsrc-for:17
2 anzsrc-for:1701
3 schema:author N888ac03d75d34011916bc38de31f8202
4 schema:datePublished 2012
5 schema:datePublishedReg 2012-01-01
6 schema:description Popular visual representations like SIFT have shown broad applicability across many task. This great generality comes naturally with a lack of specificity when focusing on a particular task or a set of classes. Metric learning approaches have been proposed to tailor general purpose representations to the needs of more specific tasks and have shown strong improvements on visual matching and recognition benchmarks. However, the performance of metric learning depends strongly on the labels that are used for learning. Therefore, we propose to combine metric learning with an active sample selection strategy in order to find labels that are representative for each class as well as improve the class separation of the learnt metric. We analyze several active sample selection strategies in terms of exploration and exploitation trade-offs. Our novel scheme achieves on three different datasets up to 10% improvement of the learned metric. We compare a batch version of our scheme to an interleaved execution of sample selection and metric learning which leads to an overall improvement of up to 23% on challenging datasets for object class recognition.
7 schema:editor Ne69e47b438664e8898485f1146833017
8 schema:genre chapter
9 schema:isAccessibleForFree false
10 schema:isPartOf N0641d944520f4192b6d03e5bc4fd70f5
11 schema:keywords SIFT
12 applicability
13 approach
14 batch version
15 benchmarks
16 broad applicability
17 class
18 class recognition
19 class separation
20 dataset
21 different datasets
22 execution
23 exploitation
24 exploration
25 general-purpose representation
26 generality
27 greater generality
28 improvement
29 interleaved execution
30 labels
31 lack
32 lack of specificity
33 learning
34 learning approach
35 matching
36 metric learning
37 metric learning approach
38 metrics
39 need
40 novel scheme
41 object class recognition
42 object recognition
43 order
44 overall improvement
45 particular task
46 performance
47 popular visual representation
48 recognition
49 recognition benchmarks
50 representation
51 sample selection
52 sample selection strategy
53 scheme
54 selection
55 selection strategy
56 separation
57 set
58 set of classes
59 specific tasks
60 specificity
61 strategies
62 strong improvement
63 task
64 terms
65 terms of exploration
66 version
67 visual matching
68 visual representation
69 schema:name Active Metric Learning for Object Recognition
70 schema:pagination 327-336
71 schema:productId N3d77afaf7bdf4dbbb589644379b7f336
72 N6b6eb4a1c9784da19de70d1c1dbf31ec
73 schema:publisher Ndba3c63a1a0c46ebab51dd9acf4c6296
74 schema:sameAs https://app.dimensions.ai/details/publication/pub.1026458093
75 https://doi.org/10.1007/978-3-642-32717-9_33
76 schema:sdDatePublished 2022-12-01T06:50
77 schema:sdLicense https://scigraph.springernature.com/explorer/license/
78 schema:sdPublisher Na1f609e84fce4706874f8ad9373e5db6
79 schema:url https://doi.org/10.1007/978-3-642-32717-9_33
80 sgo:license sg:explorer/license/
81 sgo:sdDataset chapters
82 rdf:type schema:Chapter
83 N0641d944520f4192b6d03e5bc4fd70f5 schema:isbn 978-3-642-32716-2
84 978-3-642-32717-9
85 schema:name Pattern Recognition
86 rdf:type schema:Book
87 N352e28254d374377bb6bcc567baf300c rdf:first sg:person.013361072755.17
88 rdf:rest N685ade2c6cac42c3837c2e4327747d3c
89 N3d77afaf7bdf4dbbb589644379b7f336 schema:name doi
90 schema:value 10.1007/978-3-642-32717-9_33
91 rdf:type schema:PropertyValue
92 N4d4696e409704fe7ab01c5423e16593b rdf:first N7e716ada2b874cff96b76d12a3a77a34
93 rdf:rest Ndb8e940206074beda121c071e886d4a5
94 N685ade2c6cac42c3837c2e4327747d3c rdf:first sg:person.01174260421.90
95 rdf:rest rdf:nil
96 N6b6eb4a1c9784da19de70d1c1dbf31ec schema:name dimensions_id
97 schema:value pub.1026458093
98 rdf:type schema:PropertyValue
99 N7e716ada2b874cff96b76d12a3a77a34 schema:familyName Pock
100 schema:givenName Thomas
101 rdf:type schema:Person
102 N888ac03d75d34011916bc38de31f8202 rdf:first sg:person.011333635343.49
103 rdf:rest N352e28254d374377bb6bcc567baf300c
104 Na1f609e84fce4706874f8ad9373e5db6 schema:name Springer Nature - SN SciGraph project
105 rdf:type schema:Organization
106 Na9cd2197fee1485c88e8cb964ed7c0aa schema:familyName Pinz
107 schema:givenName Axel
108 rdf:type schema:Person
109 Nb0cc8cc2a8ca4600b198949a50e51046 rdf:first Ncac2fe27e928433e94776dc8b4cca0a7
110 rdf:rest rdf:nil
111 Ncac2fe27e928433e94776dc8b4cca0a7 schema:familyName Leberl
112 schema:givenName Franz
113 rdf:type schema:Person
114 Ncc85e9a7fefe4869a43b0651dfd89a3a schema:familyName Bischof
115 schema:givenName Horst
116 rdf:type schema:Person
117 Ndb8e940206074beda121c071e886d4a5 rdf:first Ncc85e9a7fefe4869a43b0651dfd89a3a
118 rdf:rest Nb0cc8cc2a8ca4600b198949a50e51046
119 Ndba3c63a1a0c46ebab51dd9acf4c6296 schema:name Springer Nature
120 rdf:type schema:Organisation
121 Ne69e47b438664e8898485f1146833017 rdf:first Na9cd2197fee1485c88e8cb964ed7c0aa
122 rdf:rest N4d4696e409704fe7ab01c5423e16593b
123 anzsrc-for:17 schema:inDefinedTermSet anzsrc-for:
124 schema:name Psychology and Cognitive Sciences
125 rdf:type schema:DefinedTerm
126 anzsrc-for:1701 schema:inDefinedTermSet anzsrc-for:
127 schema:name Psychology
128 rdf:type schema:DefinedTerm
129 sg:person.011333635343.49 schema:affiliation grid-institutes:grid.419528.3
130 schema:familyName Ebert
131 schema:givenName Sandra
132 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.011333635343.49
133 rdf:type schema:Person
134 sg:person.01174260421.90 schema:affiliation grid-institutes:grid.419528.3
135 schema:familyName Schiele
136 schema:givenName Bernt
137 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01174260421.90
138 rdf:type schema:Person
139 sg:person.013361072755.17 schema:affiliation grid-institutes:grid.419528.3
140 schema:familyName Fritz
141 schema:givenName Mario
142 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.013361072755.17
143 rdf:type schema:Person
144 grid-institutes:grid.419528.3 schema:alternateName Max Planck Institute for Informatics, Saarbrucken, Germany
145 schema:name Max Planck Institute for Informatics, Saarbrucken, Germany
146 rdf:type schema:Organization
 




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


...