Local Features and Kernels for Classification of Texture and Object Categories: A Comprehensive Study View Full Text


Ontology type: schema:ScholarlyArticle      Open Access: True


Article Info

DATE

2007-06

AUTHORS

J. Zhang, M. Marszałek, S. Lazebnik, C. Schmid

ABSTRACT

Recently, methods based on local image features have shown promise for texture and object recognition tasks. This paper presents a large-scale evaluation of an approach that represents images as distributions (signatures or histograms) of features extracted from a sparse set of keypoint locations and learns a Support Vector Machine classifier with kernels based on two effective measures for comparing distributions, the Earth Mover’s Distance and the χ2 distance. We first evaluate the performance of our approach with different keypoint detectors and descriptors, as well as different kernels and classifiers. We then conduct a comparative evaluation with several state-of-the-art recognition methods on four texture and five object databases. On most of these databases, our implementation exceeds the best reported results and achieves comparable performance on the rest. Finally, we investigate the influence of background correlations on recognition performance via extensive tests on the PASCAL database, for which ground-truth object localization information is available. Our experiments demonstrate that image representations based on distributions of local features are surprisingly effective for classification of texture and object images under challenging real-world conditions, including significant intra-class variations and substantial background clutter. More... »

PAGES

213-238

References to SciGraph publications

  • 2004. On the Significance of Real-World Conditions for Material Classification in COMPUTER VISION - ECCV 2004
  • 2001-06. Representing and Recognizing the Visual Appearance of Materials using Three-dimensional Textons in INTERNATIONAL JOURNAL OF COMPUTER VISION
  • 1996-02. Direct computation of shape cues using scale-adapted spatial derivative operators in INTERNATIONAL JOURNAL OF COMPUTER VISION
  • 1998-11. Feature Detection with Automatic Scale Selection in INTERNATIONAL JOURNAL OF COMPUTER VISION
  • 2004-11. Distinctive Image Features from Scale-Invariant Keypoints in INTERNATIONAL JOURNAL OF COMPUTER VISION
  • 1981-03. Textons, the elements of texture perception, and their interactions in NATURE
  • 2002. Classifying Images of Materials: Achieving Viewpoint and Illumination Independence in COMPUTER VISION — ECCV 2002
  • 2002. Learning a Sparse Representation for Object Detection in COMPUTER VISION — ECCV 2002
  • 2002. An Affine Invariant Interest Point Detector in COMPUTER VISION — ECCV 2002
  • 2004. Weak Hypotheses and Boosting for Generic Object Detection and Recognition in COMPUTER VISION - ECCV 2004
  • 2004-10. Scale & Affine Invariant Interest Point Detectors in INTERNATIONAL JOURNAL OF COMPUTER VISION
  • 2000-11. The Earth Mover's Distance as a Metric for Image Retrieval in INTERNATIONAL JOURNAL OF COMPUTER VISION
  • 2005. Improving a Discriminative Approach to Object Recognition Using Image Patches in PATTERN RECOGNITION
  • 2000-01. Recognition without Correspondence using Multidimensional Receptive Field Histograms in INTERNATIONAL JOURNAL OF COMPUTER VISION
  • 2005-01. Pictorial Structures for Object Recognition in INTERNATIONAL JOURNAL OF COMPUTER VISION
  • Identifiers

    URI

    http://scigraph.springernature.com/pub.10.1007/s11263-006-9794-4

    DOI

    http://dx.doi.org/10.1007/s11263-006-9794-4

    DIMENSIONS

    https://app.dimensions.ai/details/publication/pub.1008205152


    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": {
              "name": [
                "INRIA, GRAVIR-CNRS, 655, av. de l\u2019Europe, 38330, Montbonnot, France"
              ], 
              "type": "Organization"
            }, 
            "familyName": "Zhang", 
            "givenName": "J.", 
            "id": "sg:person.01243311462.17", 
            "sameAs": [
              "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01243311462.17"
            ], 
            "type": "Person"
          }, 
          {
            "affiliation": {
              "name": [
                "INRIA, GRAVIR-CNRS, 655, av. de l\u2019Europe, 38330, Montbonnot, France"
              ], 
              "type": "Organization"
            }, 
            "familyName": "Marsza\u0142ek", 
            "givenName": "M.", 
            "id": "sg:person.011725725543.08", 
            "sameAs": [
              "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.011725725543.08"
            ], 
            "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, 405 N. Mathews Ave., 61801, Urbana, IL, USA"
              ], 
              "type": "Organization"
            }, 
            "familyName": "Lazebnik", 
            "givenName": "S.", 
            "id": "sg:person.01270671277.61", 
            "sameAs": [
              "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01270671277.61"
            ], 
            "type": "Person"
          }, 
          {
            "affiliation": {
              "name": [
                "INRIA, GRAVIR-CNRS, 655, av. de l\u2019Europe, 38330, Montbonnot, France"
              ], 
              "type": "Organization"
            }, 
            "familyName": "Schmid", 
            "givenName": "C.", 
            "id": "sg:person.014001416002.35", 
            "sameAs": [
              "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.014001416002.35"
            ], 
            "type": "Person"
          }
        ], 
        "citation": [
          {
            "id": "sg:pub.10.1023/a:1026543900054", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1006187360", 
              "https://doi.org/10.1023/a:1026543900054"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1023/a:1008120406972", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1007590088", 
              "https://doi.org/10.1023/a:1008120406972"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "https://doi.org/10.1117/12.143648", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1015213336"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1007/3-540-47979-1_8", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1015337224", 
              "https://doi.org/10.1007/3-540-47979-1_8"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1007/3-540-47977-5_17", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1016201178", 
              "https://doi.org/10.1007/3-540-47977-5_17"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "https://doi.org/10.1109/tnn.2005.848998", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1018563512"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1007/11550518_41", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1021525040", 
              "https://doi.org/10.1007/11550518_41"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1007/11550518_41", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1021525040", 
              "https://doi.org/10.1007/11550518_41"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1023/b:visi.0000042934.15159.49", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1022591804", 
              "https://doi.org/10.1023/b:visi.0000042934.15159.49"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1023/b:visi.0000027790.02288.f2", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1024638466", 
              "https://doi.org/10.1023/b:visi.0000027790.02288.f2"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1007/bf00058750", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1025865092", 
              "https://doi.org/10.1007/bf00058750"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "https://doi.org/10.1016/0031-3203(92)90099-5", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1027002452"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "https://doi.org/10.1016/0031-3203(92)90099-5", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1027002452"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1038/290091a0", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1040887630", 
              "https://doi.org/10.1038/290091a0"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1023/a:1008045108935", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1041767492", 
              "https://doi.org/10.1023/a:1008045108935"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "https://doi.org/10.1145/300776.300778", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1043167095"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1007/978-3-540-24673-2_21", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1045827469", 
              "https://doi.org/10.1007/978-3-540-24673-2_21"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1007/978-3-540-24673-2_21", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1045827469", 
              "https://doi.org/10.1007/978-3-540-24673-2_21"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1023/a:1011126920638", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1046312359", 
              "https://doi.org/10.1023/a:1011126920638"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1007/978-3-540-24671-8_6", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1046580260", 
              "https://doi.org/10.1007/978-3-540-24671-8_6"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1007/978-3-540-24671-8_6", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1046580260", 
              "https://doi.org/10.1007/978-3-540-24671-8_6"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1007/3-540-47969-4_9", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1046596731", 
              "https://doi.org/10.1007/3-540-47969-4_9"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1023/b:visi.0000029664.99615.94", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1052687286", 
              "https://doi.org/10.1023/b:visi.0000029664.99615.94"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "https://doi.org/10.1109/34.531803", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1061156442"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "https://doi.org/10.1109/34.67648", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1061156772"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "https://doi.org/10.1109/34.683777", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1061156825"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "https://doi.org/10.1109/34.765655", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1061156951"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "https://doi.org/10.1109/72.788646", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1061219239"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "https://doi.org/10.1109/t-c.1973.223602", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1061455715"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "https://doi.org/10.1109/tpami.2004.1262185", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1061742651"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "https://doi.org/10.1109/tpami.2005.151", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1061742816"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "https://doi.org/10.1109/tpami.2005.188", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1061742845"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "https://doi.org/10.1109/cvpr.2005.223", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1093271322"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "https://doi.org/10.1109/cvpr.2003.1211497", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1093305113"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "https://doi.org/10.1109/iccv.2005.66", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1093344916"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "https://doi.org/10.1109/cvpr.2003.1211479", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1093624919"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "https://doi.org/10.1109/cvpr.2005.138", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1094013418"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "https://doi.org/10.1109/iccv.2005.77", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1094132829"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "https://doi.org/10.1109/cvpr.2004.383", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1094251884"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "https://doi.org/10.1109/iccv.2005.152", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1094301320"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "https://doi.org/10.1109/icpr.2004.1334079", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1094584594"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "https://doi.org/10.1109/cvpr.2005.320", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1094611604"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "https://doi.org/10.1109/cvpr.2005.134", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1094706338"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "https://doi.org/10.1109/cvpr.2005.134", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1094706338"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "https://doi.org/10.1109/iccv.2003.1238663", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1094978467"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "https://doi.org/10.1109/icip.2003.1247092", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1095114883"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "https://doi.org/10.1109/iccv.2003.1238351", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1095214097"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "https://doi.org/10.1109/cvpr.2005.16", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1095244523"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "https://doi.org/10.1109/cvpr.2005.16", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1095244523"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "https://doi.org/10.1109/iccv.2005.239", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1095611654"
            ], 
            "type": "CreativeWork"
          }
        ], 
        "datePublished": "2007-06", 
        "datePublishedReg": "2007-06-01", 
        "description": "Recently, methods based on local image features have shown promise for texture and object recognition tasks. This paper presents a large-scale evaluation of an approach that represents images as distributions (signatures or histograms) of features extracted from a sparse set of keypoint locations and learns a Support Vector Machine classifier with kernels based on two effective measures for comparing distributions, the Earth Mover\u2019s Distance and the \u03c72 distance. We first evaluate the performance of our approach with different keypoint detectors and descriptors, as well as different kernels and classifiers. We then conduct a comparative evaluation with several state-of-the-art recognition methods on four texture and five object databases. On most of these databases, our implementation exceeds the best reported results and achieves comparable performance on the rest. Finally, we investigate the influence of background correlations on recognition performance via extensive tests on the PASCAL database, for which ground-truth object localization information is available. Our experiments demonstrate that image representations based on distributions of local features are surprisingly effective for classification of texture and object images under challenging real-world conditions, including significant intra-class variations and substantial background clutter.", 
        "genre": "research_article", 
        "id": "sg:pub.10.1007/s11263-006-9794-4", 
        "inLanguage": [
          "en"
        ], 
        "isAccessibleForFree": true, 
        "isPartOf": [
          {
            "id": "sg:journal.1032807", 
            "issn": [
              "0920-5691", 
              "1573-1405"
            ], 
            "name": "International Journal of Computer Vision", 
            "type": "Periodical"
          }, 
          {
            "issueNumber": "2", 
            "type": "PublicationIssue"
          }, 
          {
            "type": "PublicationVolume", 
            "volumeNumber": "73"
          }
        ], 
        "name": "Local Features and Kernels for Classification of Texture and Object Categories: A Comprehensive Study", 
        "pagination": "213-238", 
        "productId": [
          {
            "name": "readcube_id", 
            "type": "PropertyValue", 
            "value": [
              "2717408ec39caee10bc07b9ec1c898da906c9cc62bcb0627b2af57cd84d54ba2"
            ]
          }, 
          {
            "name": "doi", 
            "type": "PropertyValue", 
            "value": [
              "10.1007/s11263-006-9794-4"
            ]
          }, 
          {
            "name": "dimensions_id", 
            "type": "PropertyValue", 
            "value": [
              "pub.1008205152"
            ]
          }
        ], 
        "sameAs": [
          "https://doi.org/10.1007/s11263-006-9794-4", 
          "https://app.dimensions.ai/details/publication/pub.1008205152"
        ], 
        "sdDataset": "articles", 
        "sdDatePublished": "2019-04-10T13:19", 
        "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/v90/0000000001_0000000264/records_8659_00000520.jsonl", 
        "type": "ScholarlyArticle", 
        "url": "http://link.springer.com/10.1007%2Fs11263-006-9794-4"
      }
    ]
     

    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/s11263-006-9794-4'

    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/s11263-006-9794-4'

    Turtle is a human-readable linked data format.

    curl -H 'Accept: text/turtle' 'https://scigraph.springernature.com/pub.10.1007/s11263-006-9794-4'

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

    curl -H 'Accept: application/rdf+xml' 'https://scigraph.springernature.com/pub.10.1007/s11263-006-9794-4'


     

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

    235 TRIPLES      21 PREDICATES      71 URIs      19 LITERALS      7 BLANK NODES

    Subject Predicate Object
    1 sg:pub.10.1007/s11263-006-9794-4 schema:about anzsrc-for:08
    2 anzsrc-for:0801
    3 schema:author N9f0e76abd87947ce9856575c9074533f
    4 schema:citation sg:pub.10.1007/11550518_41
    5 sg:pub.10.1007/3-540-47969-4_9
    6 sg:pub.10.1007/3-540-47977-5_17
    7 sg:pub.10.1007/3-540-47979-1_8
    8 sg:pub.10.1007/978-3-540-24671-8_6
    9 sg:pub.10.1007/978-3-540-24673-2_21
    10 sg:pub.10.1007/bf00058750
    11 sg:pub.10.1023/a:1008045108935
    12 sg:pub.10.1023/a:1008120406972
    13 sg:pub.10.1023/a:1011126920638
    14 sg:pub.10.1023/a:1026543900054
    15 sg:pub.10.1023/b:visi.0000027790.02288.f2
    16 sg:pub.10.1023/b:visi.0000029664.99615.94
    17 sg:pub.10.1023/b:visi.0000042934.15159.49
    18 sg:pub.10.1038/290091a0
    19 https://doi.org/10.1016/0031-3203(92)90099-5
    20 https://doi.org/10.1109/34.531803
    21 https://doi.org/10.1109/34.67648
    22 https://doi.org/10.1109/34.683777
    23 https://doi.org/10.1109/34.765655
    24 https://doi.org/10.1109/72.788646
    25 https://doi.org/10.1109/cvpr.2003.1211479
    26 https://doi.org/10.1109/cvpr.2003.1211497
    27 https://doi.org/10.1109/cvpr.2004.383
    28 https://doi.org/10.1109/cvpr.2005.134
    29 https://doi.org/10.1109/cvpr.2005.138
    30 https://doi.org/10.1109/cvpr.2005.16
    31 https://doi.org/10.1109/cvpr.2005.223
    32 https://doi.org/10.1109/cvpr.2005.320
    33 https://doi.org/10.1109/iccv.2003.1238351
    34 https://doi.org/10.1109/iccv.2003.1238663
    35 https://doi.org/10.1109/iccv.2005.152
    36 https://doi.org/10.1109/iccv.2005.239
    37 https://doi.org/10.1109/iccv.2005.66
    38 https://doi.org/10.1109/iccv.2005.77
    39 https://doi.org/10.1109/icip.2003.1247092
    40 https://doi.org/10.1109/icpr.2004.1334079
    41 https://doi.org/10.1109/t-c.1973.223602
    42 https://doi.org/10.1109/tnn.2005.848998
    43 https://doi.org/10.1109/tpami.2004.1262185
    44 https://doi.org/10.1109/tpami.2005.151
    45 https://doi.org/10.1109/tpami.2005.188
    46 https://doi.org/10.1117/12.143648
    47 https://doi.org/10.1145/300776.300778
    48 schema:datePublished 2007-06
    49 schema:datePublishedReg 2007-06-01
    50 schema:description Recently, methods based on local image features have shown promise for texture and object recognition tasks. This paper presents a large-scale evaluation of an approach that represents images as distributions (signatures or histograms) of features extracted from a sparse set of keypoint locations and learns a Support Vector Machine classifier with kernels based on two effective measures for comparing distributions, the Earth Mover’s Distance and the χ2 distance. We first evaluate the performance of our approach with different keypoint detectors and descriptors, as well as different kernels and classifiers. We then conduct a comparative evaluation with several state-of-the-art recognition methods on four texture and five object databases. On most of these databases, our implementation exceeds the best reported results and achieves comparable performance on the rest. Finally, we investigate the influence of background correlations on recognition performance via extensive tests on the PASCAL database, for which ground-truth object localization information is available. Our experiments demonstrate that image representations based on distributions of local features are surprisingly effective for classification of texture and object images under challenging real-world conditions, including significant intra-class variations and substantial background clutter.
    51 schema:genre research_article
    52 schema:inLanguage en
    53 schema:isAccessibleForFree true
    54 schema:isPartOf N2effd30cac6a4e61bc8f893c7b310252
    55 Ne399a70282af474cb2e873a81c8e157e
    56 sg:journal.1032807
    57 schema:name Local Features and Kernels for Classification of Texture and Object Categories: A Comprehensive Study
    58 schema:pagination 213-238
    59 schema:productId N28507f33dce541bb88f66890fdfa9e56
    60 Na19abd097cf24cd2b69ecc05a0a321a7
    61 Naec99b2eb8fd413180fba938d8cd8250
    62 schema:sameAs https://app.dimensions.ai/details/publication/pub.1008205152
    63 https://doi.org/10.1007/s11263-006-9794-4
    64 schema:sdDatePublished 2019-04-10T13:19
    65 schema:sdLicense https://scigraph.springernature.com/explorer/license/
    66 schema:sdPublisher Nfb7f48be099146c2a9da2815ad729ac6
    67 schema:url http://link.springer.com/10.1007%2Fs11263-006-9794-4
    68 sgo:license sg:explorer/license/
    69 sgo:sdDataset articles
    70 rdf:type schema:ScholarlyArticle
    71 N0cf9b7855f2441a2a68535f023da8a71 schema:name INRIA, GRAVIR-CNRS, 655, av. de l’Europe, 38330, Montbonnot, France
    72 rdf:type schema:Organization
    73 N236aa910567d4584b8b962e5aba6a423 rdf:first sg:person.011725725543.08
    74 rdf:rest N4affb4b02402445f8bcef90666fcc00c
    75 N28507f33dce541bb88f66890fdfa9e56 schema:name dimensions_id
    76 schema:value pub.1008205152
    77 rdf:type schema:PropertyValue
    78 N2effd30cac6a4e61bc8f893c7b310252 schema:volumeNumber 73
    79 rdf:type schema:PublicationVolume
    80 N4affb4b02402445f8bcef90666fcc00c rdf:first sg:person.01270671277.61
    81 rdf:rest Nd882f338768d48d4920b9b2ee5e1aca4
    82 N561fd670202a4a02b133c66441c6f80a schema:name INRIA, GRAVIR-CNRS, 655, av. de l’Europe, 38330, Montbonnot, France
    83 rdf:type schema:Organization
    84 N9f0e76abd87947ce9856575c9074533f rdf:first sg:person.01243311462.17
    85 rdf:rest N236aa910567d4584b8b962e5aba6a423
    86 Na19abd097cf24cd2b69ecc05a0a321a7 schema:name readcube_id
    87 schema:value 2717408ec39caee10bc07b9ec1c898da906c9cc62bcb0627b2af57cd84d54ba2
    88 rdf:type schema:PropertyValue
    89 Naec99b2eb8fd413180fba938d8cd8250 schema:name doi
    90 schema:value 10.1007/s11263-006-9794-4
    91 rdf:type schema:PropertyValue
    92 Nd882f338768d48d4920b9b2ee5e1aca4 rdf:first sg:person.014001416002.35
    93 rdf:rest rdf:nil
    94 Ne22a40ab364442b3a7483f218be52a3c schema:name INRIA, GRAVIR-CNRS, 655, av. de l’Europe, 38330, Montbonnot, France
    95 rdf:type schema:Organization
    96 Ne399a70282af474cb2e873a81c8e157e schema:issueNumber 2
    97 rdf:type schema:PublicationIssue
    98 Nfb7f48be099146c2a9da2815ad729ac6 schema:name Springer Nature - SN SciGraph project
    99 rdf:type schema:Organization
    100 anzsrc-for:08 schema:inDefinedTermSet anzsrc-for:
    101 schema:name Information and Computing Sciences
    102 rdf:type schema:DefinedTerm
    103 anzsrc-for:0801 schema:inDefinedTermSet anzsrc-for:
    104 schema:name Artificial Intelligence and Image Processing
    105 rdf:type schema:DefinedTerm
    106 sg:journal.1032807 schema:issn 0920-5691
    107 1573-1405
    108 schema:name International Journal of Computer Vision
    109 rdf:type schema:Periodical
    110 sg:person.011725725543.08 schema:affiliation N0cf9b7855f2441a2a68535f023da8a71
    111 schema:familyName Marszałek
    112 schema:givenName M.
    113 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.011725725543.08
    114 rdf:type schema:Person
    115 sg:person.01243311462.17 schema:affiliation Ne22a40ab364442b3a7483f218be52a3c
    116 schema:familyName Zhang
    117 schema:givenName J.
    118 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01243311462.17
    119 rdf:type schema:Person
    120 sg:person.01270671277.61 schema:affiliation https://www.grid.ac/institutes/grid.35403.31
    121 schema:familyName Lazebnik
    122 schema:givenName S.
    123 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01270671277.61
    124 rdf:type schema:Person
    125 sg:person.014001416002.35 schema:affiliation N561fd670202a4a02b133c66441c6f80a
    126 schema:familyName Schmid
    127 schema:givenName C.
    128 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.014001416002.35
    129 rdf:type schema:Person
    130 sg:pub.10.1007/11550518_41 schema:sameAs https://app.dimensions.ai/details/publication/pub.1021525040
    131 https://doi.org/10.1007/11550518_41
    132 rdf:type schema:CreativeWork
    133 sg:pub.10.1007/3-540-47969-4_9 schema:sameAs https://app.dimensions.ai/details/publication/pub.1046596731
    134 https://doi.org/10.1007/3-540-47969-4_9
    135 rdf:type schema:CreativeWork
    136 sg:pub.10.1007/3-540-47977-5_17 schema:sameAs https://app.dimensions.ai/details/publication/pub.1016201178
    137 https://doi.org/10.1007/3-540-47977-5_17
    138 rdf:type schema:CreativeWork
    139 sg:pub.10.1007/3-540-47979-1_8 schema:sameAs https://app.dimensions.ai/details/publication/pub.1015337224
    140 https://doi.org/10.1007/3-540-47979-1_8
    141 rdf:type schema:CreativeWork
    142 sg:pub.10.1007/978-3-540-24671-8_6 schema:sameAs https://app.dimensions.ai/details/publication/pub.1046580260
    143 https://doi.org/10.1007/978-3-540-24671-8_6
    144 rdf:type schema:CreativeWork
    145 sg:pub.10.1007/978-3-540-24673-2_21 schema:sameAs https://app.dimensions.ai/details/publication/pub.1045827469
    146 https://doi.org/10.1007/978-3-540-24673-2_21
    147 rdf:type schema:CreativeWork
    148 sg:pub.10.1007/bf00058750 schema:sameAs https://app.dimensions.ai/details/publication/pub.1025865092
    149 https://doi.org/10.1007/bf00058750
    150 rdf:type schema:CreativeWork
    151 sg:pub.10.1023/a:1008045108935 schema:sameAs https://app.dimensions.ai/details/publication/pub.1041767492
    152 https://doi.org/10.1023/a:1008045108935
    153 rdf:type schema:CreativeWork
    154 sg:pub.10.1023/a:1008120406972 schema:sameAs https://app.dimensions.ai/details/publication/pub.1007590088
    155 https://doi.org/10.1023/a:1008120406972
    156 rdf:type schema:CreativeWork
    157 sg:pub.10.1023/a:1011126920638 schema:sameAs https://app.dimensions.ai/details/publication/pub.1046312359
    158 https://doi.org/10.1023/a:1011126920638
    159 rdf:type schema:CreativeWork
    160 sg:pub.10.1023/a:1026543900054 schema:sameAs https://app.dimensions.ai/details/publication/pub.1006187360
    161 https://doi.org/10.1023/a:1026543900054
    162 rdf:type schema:CreativeWork
    163 sg:pub.10.1023/b:visi.0000027790.02288.f2 schema:sameAs https://app.dimensions.ai/details/publication/pub.1024638466
    164 https://doi.org/10.1023/b:visi.0000027790.02288.f2
    165 rdf:type schema:CreativeWork
    166 sg:pub.10.1023/b:visi.0000029664.99615.94 schema:sameAs https://app.dimensions.ai/details/publication/pub.1052687286
    167 https://doi.org/10.1023/b:visi.0000029664.99615.94
    168 rdf:type schema:CreativeWork
    169 sg:pub.10.1023/b:visi.0000042934.15159.49 schema:sameAs https://app.dimensions.ai/details/publication/pub.1022591804
    170 https://doi.org/10.1023/b:visi.0000042934.15159.49
    171 rdf:type schema:CreativeWork
    172 sg:pub.10.1038/290091a0 schema:sameAs https://app.dimensions.ai/details/publication/pub.1040887630
    173 https://doi.org/10.1038/290091a0
    174 rdf:type schema:CreativeWork
    175 https://doi.org/10.1016/0031-3203(92)90099-5 schema:sameAs https://app.dimensions.ai/details/publication/pub.1027002452
    176 rdf:type schema:CreativeWork
    177 https://doi.org/10.1109/34.531803 schema:sameAs https://app.dimensions.ai/details/publication/pub.1061156442
    178 rdf:type schema:CreativeWork
    179 https://doi.org/10.1109/34.67648 schema:sameAs https://app.dimensions.ai/details/publication/pub.1061156772
    180 rdf:type schema:CreativeWork
    181 https://doi.org/10.1109/34.683777 schema:sameAs https://app.dimensions.ai/details/publication/pub.1061156825
    182 rdf:type schema:CreativeWork
    183 https://doi.org/10.1109/34.765655 schema:sameAs https://app.dimensions.ai/details/publication/pub.1061156951
    184 rdf:type schema:CreativeWork
    185 https://doi.org/10.1109/72.788646 schema:sameAs https://app.dimensions.ai/details/publication/pub.1061219239
    186 rdf:type schema:CreativeWork
    187 https://doi.org/10.1109/cvpr.2003.1211479 schema:sameAs https://app.dimensions.ai/details/publication/pub.1093624919
    188 rdf:type schema:CreativeWork
    189 https://doi.org/10.1109/cvpr.2003.1211497 schema:sameAs https://app.dimensions.ai/details/publication/pub.1093305113
    190 rdf:type schema:CreativeWork
    191 https://doi.org/10.1109/cvpr.2004.383 schema:sameAs https://app.dimensions.ai/details/publication/pub.1094251884
    192 rdf:type schema:CreativeWork
    193 https://doi.org/10.1109/cvpr.2005.134 schema:sameAs https://app.dimensions.ai/details/publication/pub.1094706338
    194 rdf:type schema:CreativeWork
    195 https://doi.org/10.1109/cvpr.2005.138 schema:sameAs https://app.dimensions.ai/details/publication/pub.1094013418
    196 rdf:type schema:CreativeWork
    197 https://doi.org/10.1109/cvpr.2005.16 schema:sameAs https://app.dimensions.ai/details/publication/pub.1095244523
    198 rdf:type schema:CreativeWork
    199 https://doi.org/10.1109/cvpr.2005.223 schema:sameAs https://app.dimensions.ai/details/publication/pub.1093271322
    200 rdf:type schema:CreativeWork
    201 https://doi.org/10.1109/cvpr.2005.320 schema:sameAs https://app.dimensions.ai/details/publication/pub.1094611604
    202 rdf:type schema:CreativeWork
    203 https://doi.org/10.1109/iccv.2003.1238351 schema:sameAs https://app.dimensions.ai/details/publication/pub.1095214097
    204 rdf:type schema:CreativeWork
    205 https://doi.org/10.1109/iccv.2003.1238663 schema:sameAs https://app.dimensions.ai/details/publication/pub.1094978467
    206 rdf:type schema:CreativeWork
    207 https://doi.org/10.1109/iccv.2005.152 schema:sameAs https://app.dimensions.ai/details/publication/pub.1094301320
    208 rdf:type schema:CreativeWork
    209 https://doi.org/10.1109/iccv.2005.239 schema:sameAs https://app.dimensions.ai/details/publication/pub.1095611654
    210 rdf:type schema:CreativeWork
    211 https://doi.org/10.1109/iccv.2005.66 schema:sameAs https://app.dimensions.ai/details/publication/pub.1093344916
    212 rdf:type schema:CreativeWork
    213 https://doi.org/10.1109/iccv.2005.77 schema:sameAs https://app.dimensions.ai/details/publication/pub.1094132829
    214 rdf:type schema:CreativeWork
    215 https://doi.org/10.1109/icip.2003.1247092 schema:sameAs https://app.dimensions.ai/details/publication/pub.1095114883
    216 rdf:type schema:CreativeWork
    217 https://doi.org/10.1109/icpr.2004.1334079 schema:sameAs https://app.dimensions.ai/details/publication/pub.1094584594
    218 rdf:type schema:CreativeWork
    219 https://doi.org/10.1109/t-c.1973.223602 schema:sameAs https://app.dimensions.ai/details/publication/pub.1061455715
    220 rdf:type schema:CreativeWork
    221 https://doi.org/10.1109/tnn.2005.848998 schema:sameAs https://app.dimensions.ai/details/publication/pub.1018563512
    222 rdf:type schema:CreativeWork
    223 https://doi.org/10.1109/tpami.2004.1262185 schema:sameAs https://app.dimensions.ai/details/publication/pub.1061742651
    224 rdf:type schema:CreativeWork
    225 https://doi.org/10.1109/tpami.2005.151 schema:sameAs https://app.dimensions.ai/details/publication/pub.1061742816
    226 rdf:type schema:CreativeWork
    227 https://doi.org/10.1109/tpami.2005.188 schema:sameAs https://app.dimensions.ai/details/publication/pub.1061742845
    228 rdf:type schema:CreativeWork
    229 https://doi.org/10.1117/12.143648 schema:sameAs https://app.dimensions.ai/details/publication/pub.1015213336
    230 rdf:type schema:CreativeWork
    231 https://doi.org/10.1145/300776.300778 schema:sameAs https://app.dimensions.ai/details/publication/pub.1043167095
    232 rdf:type schema:CreativeWork
    233 https://www.grid.ac/institutes/grid.35403.31 schema:alternateName University of Illinois at Urbana Champaign
    234 schema:name Beckman Institute, University of Illinois, 405 N. Mathews Ave., 61801, Urbana, IL, USA
    235 rdf:type schema:Organization
     




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


    ...