A New Class of Learnable Detectors for Categorisation
Identifieur interne : 001356 ( Main/Exploration ); précédent : 001355; suivant : 001357A New Class of Learnable Detectors for Categorisation
Auteurs : Jiri Matas [République tchèque] ; Karel Zimmermann [République tchèque]Source :
- Lecture Notes in Computer Science [ 0302-9743 ] ; 2005.
Abstract
Abstract: A new class of image-level detectors that can be adapted by machine learning techniques to detect parts of objects from a given category is proposed. A classifier (e.g. neural network or adaboost trained classifier) within the detector selects a relevant subset of extremal regions, i.e. regions that are connected components of a thresholded image. Properties of extremal regions render the detector very robust to illumination change. Robustness to viewpoint change is achieved by using invariant descriptors and/or by modeling shape variations by the classifier. The approach is brought to bear on three problems: text detection, face segmentation and leopard skin detection. High detection rates were obtained for unconstrained (i.e. brightness, affine and font invariant) text detection (92%) with a reasonable false positive rate. The time-complexity of the detection is approximately linear in the number of pixels and a non-optimized implementation runs at about 1 frame per second for a 640× 480 image on a high-end PC.
Url:
DOI: 10.1007/11499145_55
Affiliations:
Links toward previous steps (curation, corpus...)
- to stream Istex, to step Corpus: 002F78
- to stream Istex, to step Curation: 002D45
- to stream Istex, to step Checkpoint: 000C65
- to stream Main, to step Merge: 001392
- to stream Main, to step Curation: 001356
Le document en format XML
<record><TEI wicri:istexFullTextTei="biblStruct"><teiHeader><fileDesc><titleStmt><title xml:lang="en">A New Class of Learnable Detectors for Categorisation</title><author><name sortKey="Matas, Jiri" sort="Matas, Jiri" uniqKey="Matas J" first="Jiri" last="Matas">Jiri Matas</name></author><author><name sortKey="Zimmermann, Karel" sort="Zimmermann, Karel" uniqKey="Zimmermann K" first="Karel" last="Zimmermann">Karel Zimmermann</name></author></titleStmt><publicationStmt><idno type="wicri:source">ISTEX</idno><idno type="RBID">ISTEX:BFE094E4701A0F3D0CAEAEB1C633B3A601A21A13</idno><date when="2005" year="2005">2005</date><idno type="doi">10.1007/11499145_55</idno><idno type="url">https://api.istex.fr/document/BFE094E4701A0F3D0CAEAEB1C633B3A601A21A13/fulltext/pdf</idno><idno type="wicri:Area/Istex/Corpus">002F78</idno><idno type="wicri:Area/Istex/Curation">002D45</idno><idno type="wicri:Area/Istex/Checkpoint">000C65</idno><idno type="wicri:doubleKey">0302-9743:2005:Matas J:a:new:class</idno><idno type="wicri:Area/Main/Merge">001392</idno><idno type="wicri:Area/Main/Curation">001356</idno><idno type="wicri:Area/Main/Exploration">001356</idno></publicationStmt><sourceDesc><biblStruct><analytic><title level="a" type="main" xml:lang="en">A New Class of Learnable Detectors for Categorisation</title><author><name sortKey="Matas, Jiri" sort="Matas, Jiri" uniqKey="Matas J" first="Jiri" last="Matas">Jiri Matas</name><affiliation wicri:level="3"><country>République tchèque</country><placeName><settlement type="city">Prague</settlement><region type="région" nuts="2">Bohême centrale</region></placeName><wicri:orgArea>Center for Machine Perception, Faculty of Electrotechnical Engineering</wicri:orgArea></affiliation></author><author><name sortKey="Zimmermann, Karel" sort="Zimmermann, Karel" uniqKey="Zimmermann K" first="Karel" last="Zimmermann">Karel Zimmermann</name><affiliation wicri:level="3"><country>République tchèque</country><placeName><settlement type="city">Prague</settlement><region type="région" nuts="2">Bohême centrale</region></placeName><wicri:orgArea>Center for Machine Perception, Faculty of Electrotechnical Engineering</wicri:orgArea></affiliation></author></analytic><monogr></monogr><series><title level="s">Lecture Notes in Computer Science</title><imprint><date>2005</date></imprint><idno type="ISSN">0302-9743</idno><idno type="eISSN">1611-3349</idno><idno type="ISSN">0302-9743</idno></series><idno type="istex">BFE094E4701A0F3D0CAEAEB1C633B3A601A21A13</idno><idno type="DOI">10.1007/11499145_55</idno><idno type="ChapterID">55</idno><idno type="ChapterID">Chap55</idno></biblStruct></sourceDesc><seriesStmt><idno type="ISSN">0302-9743</idno></seriesStmt></fileDesc><profileDesc><textClass></textClass><langUsage><language ident="en">en</language></langUsage></profileDesc></teiHeader><front><div type="abstract" xml:lang="en">Abstract: A new class of image-level detectors that can be adapted by machine learning techniques to detect parts of objects from a given category is proposed. A classifier (e.g. neural network or adaboost trained classifier) within the detector selects a relevant subset of extremal regions, i.e. regions that are connected components of a thresholded image. Properties of extremal regions render the detector very robust to illumination change. Robustness to viewpoint change is achieved by using invariant descriptors and/or by modeling shape variations by the classifier. The approach is brought to bear on three problems: text detection, face segmentation and leopard skin detection. High detection rates were obtained for unconstrained (i.e. brightness, affine and font invariant) text detection (92%) with a reasonable false positive rate. The time-complexity of the detection is approximately linear in the number of pixels and a non-optimized implementation runs at about 1 frame per second for a 640× 480 image on a high-end PC.</div></front></TEI><affiliations><list><country><li>République tchèque</li></country><region><li>Bohême centrale</li></region><settlement><li>Prague</li></settlement></list><tree><country name="République tchèque"><region name="Bohême centrale"><name sortKey="Matas, Jiri" sort="Matas, Jiri" uniqKey="Matas J" first="Jiri" last="Matas">Jiri Matas</name></region><name sortKey="Zimmermann, Karel" sort="Zimmermann, Karel" uniqKey="Zimmermann K" first="Karel" last="Zimmermann">Karel Zimmermann</name></country></tree></affiliations></record>Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Ticri/CIDE/explor/OcrV1/Data/Main/Exploration
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 001356 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Main/Exploration/biblio.hfd -nk 001356 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien
|wiki= Ticri/CIDE
|area= OcrV1
|flux= Main
|étape= Exploration
|type= RBID
|clé= ISTEX:BFE094E4701A0F3D0CAEAEB1C633B3A601A21A13
|texte= A New Class of Learnable Detectors for Categorisation
}}
| This area was generated with Dilib version V0.6.32. |  |