Image Progressive Retrieval from a Videodisk
Identifieur interne : 000789 ( Crin/Curation ); précédent : 000788; suivant : 000790Image Progressive Retrieval from a Videodisk
Auteurs : G. Halin ; M. CréhangeSource :
English descriptors
Abstract
You need finding images to illustrate an article or a lecture. An optical disk could be an excellent source, containing a large amount of pictures. Our subject is to propose means of aiding you in retrieving suitable images, even though you might be unable to directly formulate an adequate request. Between strict retrieval and random browsing, we study a progressive retrieval based on ``relevance feedback'' processed through a man-machine dialogue. In it, images play an inportant role, and also the knowledge the system has about the user. So it is in the EXPRIM system. In the system, we suppose that the image base (on a videodisk) has been coupled with a documentary base. The first penetration in the base is made through a textual request, or through any more or less pre-determined browsing among the image collection. The result of this penetration is a set of images. While visualizing this set, the user then chooses the images which match his need and rejects those which don't. The system next tries to build a new request which is hoped to provide a better set of images. This is done by trying to understand the user's need through his choice\, : the chosen images are positive illustrations of it while the rejected ones are negative illustrations. This may be viewed as a machine learning process by examples and negative examples, the concept to learn being the user's need.
Links toward previous steps (curation, corpus...)
- to stream Crin, to step Corpus: Pour aller vers cette notice dans l'étape Curation :000789
Links to Exploration step
CRIN:halin89aLe document en format XML
<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="fr" wicri:score="-44">Image Progressive Retrieval from a Videodisk</title>
</titleStmt>
<publicationStmt><idno type="RBID">CRIN:halin89a</idno>
<date when="1989" year="1989">1989</date>
<idno type="wicri:Area/Crin/Corpus">000789</idno>
<idno type="wicri:Area/Crin/Curation">000789</idno>
<idno type="wicri:explorRef" wicri:stream="Crin" wicri:step="Curation">000789</idno>
</publicationStmt>
<sourceDesc><biblStruct><analytic><title xml:lang="fr">Image Progressive Retrieval from a Videodisk</title>
<author><name sortKey="Halin, G" sort="Halin, G" uniqKey="Halin G" first="G." last="Halin">G. Halin</name>
</author>
<author><name sortKey="Crehange, M" sort="Crehange, M" uniqKey="Crehange M" first="M." last="Créhange">M. Créhange</name>
</author>
</analytic>
</biblStruct>
</sourceDesc>
</fileDesc>
<profileDesc><textClass><keywords scheme="KwdEn" xml:lang="en"><term>image retrieval</term>
<term>machine learning</term>
<term>man machine communication</term>
<term>videodisk</term>
</keywords>
</textClass>
</profileDesc>
</teiHeader>
<front><div type="abstract" xml:lang="en" wicri:score="2538">You need finding images to illustrate an article or a lecture. An optical disk could be an excellent source, containing a large amount of pictures. Our subject is to propose means of aiding you in retrieving suitable images, even though you might be unable to directly formulate an adequate request. Between strict retrieval and random browsing, we study a progressive retrieval based on ``relevance feedback'' processed through a man-machine dialogue. In it, images play an inportant role, and also the knowledge the system has about the user. So it is in the EXPRIM system. In the system, we suppose that the image base (on a videodisk) has been coupled with a documentary base. The first penetration in the base is made through a textual request, or through any more or less pre-determined browsing among the image collection. The result of this penetration is a set of images. While visualizing this set, the user then chooses the images which match his need and rejects those which don't. The system next tries to build a new request which is hoped to provide a better set of images. This is done by trying to understand the user's need through his choice\, : the chosen images are positive illustrations of it while the rejected ones are negative illustrations. This may be viewed as a machine learning process by examples and negative examples, the concept to learn being the user's need.</div>
</front>
</TEI>
<BibTex type="inproceedings"><ref>halin89a</ref>
<crinnumber>89-r-075</crinnumber>
<category>3</category>
<equipe>EXPRIM</equipe>
<author><e>Halin, G.</e>
<e>Créhange, M.</e>
</author>
<title>Image Progressive Retrieval from a Videodisk</title>
<booktitle>{Proceedings OPTICALINFO 89 International Meeting for Optical Publishing and Storage, Amsterdam (Pays Bas)}</booktitle>
<year>1989</year>
<pages>67-65</pages>
<month>apr</month>
<publisher>Learned Information</publisher>
<keywords><e>image retrieval</e>
<e>videodisk</e>
<e>machine learning</e>
<e>man machine communication</e>
</keywords>
<abstract>You need finding images to illustrate an article or a lecture. An optical disk could be an excellent source, containing a large amount of pictures. Our subject is to propose means of aiding you in retrieving suitable images, even though you might be unable to directly formulate an adequate request. Between strict retrieval and random browsing, we study a progressive retrieval based on ``relevance feedback'' processed through a man-machine dialogue. In it, images play an inportant role, and also the knowledge the system has about the user. So it is in the EXPRIM system. In the system, we suppose that the image base (on a videodisk) has been coupled with a documentary base. The first penetration in the base is made through a textual request, or through any more or less pre-determined browsing among the image collection. The result of this penetration is a set of images. While visualizing this set, the user then chooses the images which match his need and rejects those which don't. The system next tries to build a new request which is hoped to provide a better set of images. This is done by trying to understand the user's need through his choice\, : the chosen images are positive illustrations of it while the rejected ones are negative illustrations. This may be viewed as a machine learning process by examples and negative examples, the concept to learn being the user's need.</abstract>
</BibTex>
</record>
Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Wicri/Lorraine/explor/InforLorV4/Data/Crin/Curation
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 000789 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Crin/Curation/biblio.hfd -nk 000789 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien |wiki= Wicri/Lorraine |area= InforLorV4 |flux= Crin |étape= Curation |type= RBID |clé= CRIN:halin89a |texte= Image Progressive Retrieval from a Videodisk }}
![]() | This area was generated with Dilib version V0.6.33. | ![]() |