A Comparison between Multi-Layer Perceptrons and Convolutional Neural Networks for Text Image Super-Resolution
Identifieur interne : 000025 ( Main/Exploration ); précédent : 000024; suivant : 000026A Comparison between Multi-Layer Perceptrons and Convolutional Neural Networks for Text Image Super-Resolution
Auteurs : Clément Peyrard [France] ; Franck Mamalet [France] ; Christophe Garcia [France]Source :
English descriptors
Abstract
We compare the performances of several Multi-Layer Perceptrons (MLPs) and Convolutional Neural Networks (ConvNets) for single text image Super-Resolution. We propose an example-based framework for both MLP and ConvNet, where a non-linear mapping between pairs of patches and high-frequency pixel values is learned. We then demonstrate that for equivalent complexity, ConvNets are better than MLPs at predicting missing details in upsampled text images. To evaluate the performances, we make use of a recent database (ULR-textSISR-2013a) along with different quality measures. We show that the proposed methods outperforms sparse coding-based methods for this database.
Url:
DOI: 10.5220/0005297200840091
Affiliations:
Links toward previous steps (curation, corpus...)
- to stream Hal, to step Corpus: 000002
- to stream Hal, to step Curation: 000002
- to stream Hal, to step Checkpoint: 000013
- to stream Main, to step Merge: 000025
- to stream Main, to step Curation: 000025
Le document en format XML
<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="en">A Comparison between Multi-Layer Perceptrons and Convolutional Neural Networks for Text Image Super-Resolution</title>
<author><name sortKey="Peyrard, Clement" sort="Peyrard, Clement" uniqKey="Peyrard C" first="Clément" last="Peyrard">Clément Peyrard</name>
<affiliation wicri:level="1"><hal:affiliation type="laboratory" xml:id="struct-204598" status="VALID"><orgName>Orange Labs [Cesson-Sévigné]</orgName>
<desc><address><addrLine>4 Rue du Clos Courtel, 35510 Cesson-Sévigné</addrLine>
<country key="FR"></country>
</address>
</desc>
<listRelation><relation active="#struct-366011" type="direct"></relation>
</listRelation>
<tutelles><tutelle active="#struct-366011" type="direct"><org type="institution" xml:id="struct-366011" status="INCOMING"><orgName>Orange</orgName>
<desc><address><country key="FR"></country>
</address>
</desc>
</org>
</tutelle>
</tutelles>
</hal:affiliation>
<country>France</country>
</affiliation>
</author>
<author><name sortKey="Mamalet, Franck" sort="Mamalet, Franck" uniqKey="Mamalet F" first="Franck" last="Mamalet">Franck Mamalet</name>
<affiliation wicri:level="1"><hal:affiliation type="laboratory" xml:id="struct-204598" status="VALID"><orgName>Orange Labs [Cesson-Sévigné]</orgName>
<desc><address><addrLine>4 Rue du Clos Courtel, 35510 Cesson-Sévigné</addrLine>
<country key="FR"></country>
</address>
</desc>
<listRelation><relation active="#struct-366011" type="direct"></relation>
</listRelation>
<tutelles><tutelle active="#struct-366011" type="direct"><org type="institution" xml:id="struct-366011" status="INCOMING"><orgName>Orange</orgName>
<desc><address><country key="FR"></country>
</address>
</desc>
</org>
</tutelle>
</tutelles>
</hal:affiliation>
<country>France</country>
</affiliation>
</author>
<author><name sortKey="Garcia, Christophe" sort="Garcia, Christophe" uniqKey="Garcia C" first="Christophe" last="Garcia">Christophe Garcia</name>
<affiliation wicri:level="1"><hal:affiliation type="researchteam" xml:id="struct-403930" status="VALID"><orgName>Extraction de Caractéristiques et Identification</orgName>
<orgName type="acronym">imagine</orgName>
<desc><address><country key="FR"></country>
</address>
</desc>
<listRelation><relation active="#struct-2003" type="direct"></relation>
<relation active="#struct-33804" type="indirect"></relation>
<relation active="#struct-126765" type="indirect"></relation>
<relation active="#struct-194495" type="indirect"></relation>
<relation name="- LYON" active="#struct-301232" type="indirect"></relation>
<relation name="UMR5205" active="#struct-441569" type="indirect"></relation>
</listRelation>
<tutelles><tutelle active="#struct-2003" type="direct"><org type="laboratory" xml:id="struct-2003" status="VALID"><orgName>Laboratoire d'InfoRmatique en Image et Systèmes d'information</orgName>
<orgName type="acronym">LIRIS</orgName>
<desc><address><addrLine>Bâtiment Blaise Pascal - 20, avenue Albert Einstein - 69621 Villeurbanne cedex</addrLine>
<country key="FR"></country>
</address>
<ref type="url">http://liris.cnrs.fr/</ref>
</desc>
<listRelation><relation active="#struct-33804" type="direct"></relation>
<relation active="#struct-126765" type="direct"></relation>
<relation active="#struct-194495" type="direct"></relation>
<relation name="- LYON" active="#struct-301232" type="direct"></relation>
<relation name="UMR5205" active="#struct-441569" type="direct"></relation>
</listRelation>
</org>
</tutelle>
<tutelle active="#struct-33804" type="indirect"><org type="institution" xml:id="struct-33804" status="VALID"><orgName>Université Lumière - Lyon 2</orgName>
<orgName type="acronym">UL2</orgName>
<desc><address><addrLine>86, rue Pasteur - 69007 Lyon</addrLine>
<country key="FR"></country>
</address>
<ref type="url">http://www.univ-lyon2.fr</ref>
</desc>
</org>
</tutelle>
<tutelle active="#struct-126765" type="indirect"><org type="institution" xml:id="struct-126765" status="VALID"><orgName>École Centrale de Lyon</orgName>
<orgName type="acronym">ECL</orgName>
<desc><address><addrLine>36 avenue Guy de Collongue - 69134 Ecully cedex</addrLine>
<country key="FR"></country>
</address>
<ref type="url">http://www.ec-lyon.fr</ref>
</desc>
</org>
</tutelle>
<tutelle active="#struct-194495" type="indirect"><org type="institution" xml:id="struct-194495" status="VALID"><orgName>Université Claude Bernard Lyon 1</orgName>
<orgName type="acronym">UCBL</orgName>
<desc><address><addrLine>43, boulevard du 11 novembre 1918, 69622 Villeurbanne cedex</addrLine>
<country key="FR"></country>
</address>
<ref type="url">http://www.univ-lyon1.fr/</ref>
</desc>
</org>
</tutelle>
<tutelle name="- LYON" active="#struct-301232" type="indirect"><org type="institution" xml:id="struct-301232" status="VALID"><orgName>Institut National des Sciences Appliquées</orgName>
<orgName type="acronym">INSA</orgName>
<desc><address><country key="FR"></country>
</address>
</desc>
</org>
</tutelle>
<tutelle name="UMR5205" active="#struct-441569" type="indirect"><org type="institution" xml:id="struct-441569" status="VALID"><idno type="IdRef">02636817X</idno>
<idno type="ISNI">0000000122597504</idno>
<orgName>Centre National de la Recherche Scientifique</orgName>
<orgName type="acronym">CNRS</orgName>
<date type="start">1939-10-19</date>
<desc><address><country key="FR"></country>
</address>
<ref type="url">http://www.cnrs.fr/</ref>
</desc>
</org>
</tutelle>
</tutelles>
</hal:affiliation>
<country>France</country>
<placeName><settlement type="city">Lyon</settlement>
<region type="region" nuts="2">Auvergne-Rhône-Alpes</region>
<region type="old region" nuts="2">Rhône-Alpes</region>
</placeName>
<orgName type="university">Université Claude Bernard Lyon 1</orgName>
<orgName type="institution" wicri:auto="newGroup">Université de Lyon</orgName>
</affiliation>
</author>
</titleStmt>
<publicationStmt><idno type="wicri:source">HAL</idno>
<idno type="RBID">Hal:hal-01260671</idno>
<idno type="halId">hal-01260671</idno>
<idno type="halUri">https://hal.archives-ouvertes.fr/hal-01260671</idno>
<idno type="url">https://hal.archives-ouvertes.fr/hal-01260671</idno>
<idno type="doi">10.5220/0005297200840091</idno>
<date when="2015-03-11">2015-03-11</date>
<idno type="wicri:Area/Hal/Corpus">000002</idno>
<idno type="wicri:Area/Hal/Curation">000002</idno>
<idno type="wicri:Area/Hal/Checkpoint">000013</idno>
<idno type="wicri:Area/Main/Merge">000025</idno>
<idno type="wicri:Area/Main/Curation">000025</idno>
<idno type="wicri:Area/Main/Exploration">000025</idno>
</publicationStmt>
<sourceDesc><biblStruct><analytic><title xml:lang="en">A Comparison between Multi-Layer Perceptrons and Convolutional Neural Networks for Text Image Super-Resolution</title>
<author><name sortKey="Peyrard, Clement" sort="Peyrard, Clement" uniqKey="Peyrard C" first="Clément" last="Peyrard">Clément Peyrard</name>
<affiliation wicri:level="1"><hal:affiliation type="laboratory" xml:id="struct-204598" status="VALID"><orgName>Orange Labs [Cesson-Sévigné]</orgName>
<desc><address><addrLine>4 Rue du Clos Courtel, 35510 Cesson-Sévigné</addrLine>
<country key="FR"></country>
</address>
</desc>
<listRelation><relation active="#struct-366011" type="direct"></relation>
</listRelation>
<tutelles><tutelle active="#struct-366011" type="direct"><org type="institution" xml:id="struct-366011" status="INCOMING"><orgName>Orange</orgName>
<desc><address><country key="FR"></country>
</address>
</desc>
</org>
</tutelle>
</tutelles>
</hal:affiliation>
<country>France</country>
</affiliation>
</author>
<author><name sortKey="Mamalet, Franck" sort="Mamalet, Franck" uniqKey="Mamalet F" first="Franck" last="Mamalet">Franck Mamalet</name>
<affiliation wicri:level="1"><hal:affiliation type="laboratory" xml:id="struct-204598" status="VALID"><orgName>Orange Labs [Cesson-Sévigné]</orgName>
<desc><address><addrLine>4 Rue du Clos Courtel, 35510 Cesson-Sévigné</addrLine>
<country key="FR"></country>
</address>
</desc>
<listRelation><relation active="#struct-366011" type="direct"></relation>
</listRelation>
<tutelles><tutelle active="#struct-366011" type="direct"><org type="institution" xml:id="struct-366011" status="INCOMING"><orgName>Orange</orgName>
<desc><address><country key="FR"></country>
</address>
</desc>
</org>
</tutelle>
</tutelles>
</hal:affiliation>
<country>France</country>
</affiliation>
</author>
<author><name sortKey="Garcia, Christophe" sort="Garcia, Christophe" uniqKey="Garcia C" first="Christophe" last="Garcia">Christophe Garcia</name>
<affiliation wicri:level="1"><hal:affiliation type="researchteam" xml:id="struct-403930" status="VALID"><orgName>Extraction de Caractéristiques et Identification</orgName>
<orgName type="acronym">imagine</orgName>
<desc><address><country key="FR"></country>
</address>
</desc>
<listRelation><relation active="#struct-2003" type="direct"></relation>
<relation active="#struct-33804" type="indirect"></relation>
<relation active="#struct-126765" type="indirect"></relation>
<relation active="#struct-194495" type="indirect"></relation>
<relation name="- LYON" active="#struct-301232" type="indirect"></relation>
<relation name="UMR5205" active="#struct-441569" type="indirect"></relation>
</listRelation>
<tutelles><tutelle active="#struct-2003" type="direct"><org type="laboratory" xml:id="struct-2003" status="VALID"><orgName>Laboratoire d'InfoRmatique en Image et Systèmes d'information</orgName>
<orgName type="acronym">LIRIS</orgName>
<desc><address><addrLine>Bâtiment Blaise Pascal - 20, avenue Albert Einstein - 69621 Villeurbanne cedex</addrLine>
<country key="FR"></country>
</address>
<ref type="url">http://liris.cnrs.fr/</ref>
</desc>
<listRelation><relation active="#struct-33804" type="direct"></relation>
<relation active="#struct-126765" type="direct"></relation>
<relation active="#struct-194495" type="direct"></relation>
<relation name="- LYON" active="#struct-301232" type="direct"></relation>
<relation name="UMR5205" active="#struct-441569" type="direct"></relation>
</listRelation>
</org>
</tutelle>
<tutelle active="#struct-33804" type="indirect"><org type="institution" xml:id="struct-33804" status="VALID"><orgName>Université Lumière - Lyon 2</orgName>
<orgName type="acronym">UL2</orgName>
<desc><address><addrLine>86, rue Pasteur - 69007 Lyon</addrLine>
<country key="FR"></country>
</address>
<ref type="url">http://www.univ-lyon2.fr</ref>
</desc>
</org>
</tutelle>
<tutelle active="#struct-126765" type="indirect"><org type="institution" xml:id="struct-126765" status="VALID"><orgName>École Centrale de Lyon</orgName>
<orgName type="acronym">ECL</orgName>
<desc><address><addrLine>36 avenue Guy de Collongue - 69134 Ecully cedex</addrLine>
<country key="FR"></country>
</address>
<ref type="url">http://www.ec-lyon.fr</ref>
</desc>
</org>
</tutelle>
<tutelle active="#struct-194495" type="indirect"><org type="institution" xml:id="struct-194495" status="VALID"><orgName>Université Claude Bernard Lyon 1</orgName>
<orgName type="acronym">UCBL</orgName>
<desc><address><addrLine>43, boulevard du 11 novembre 1918, 69622 Villeurbanne cedex</addrLine>
<country key="FR"></country>
</address>
<ref type="url">http://www.univ-lyon1.fr/</ref>
</desc>
</org>
</tutelle>
<tutelle name="- LYON" active="#struct-301232" type="indirect"><org type="institution" xml:id="struct-301232" status="VALID"><orgName>Institut National des Sciences Appliquées</orgName>
<orgName type="acronym">INSA</orgName>
<desc><address><country key="FR"></country>
</address>
</desc>
</org>
</tutelle>
<tutelle name="UMR5205" active="#struct-441569" type="indirect"><org type="institution" xml:id="struct-441569" status="VALID"><idno type="IdRef">02636817X</idno>
<idno type="ISNI">0000000122597504</idno>
<orgName>Centre National de la Recherche Scientifique</orgName>
<orgName type="acronym">CNRS</orgName>
<date type="start">1939-10-19</date>
<desc><address><country key="FR"></country>
</address>
<ref type="url">http://www.cnrs.fr/</ref>
</desc>
</org>
</tutelle>
</tutelles>
</hal:affiliation>
<country>France</country>
<placeName><settlement type="city">Lyon</settlement>
<region type="region" nuts="2">Auvergne-Rhône-Alpes</region>
<region type="old region" nuts="2">Rhône-Alpes</region>
</placeName>
<orgName type="university">Université Claude Bernard Lyon 1</orgName>
<orgName type="institution" wicri:auto="newGroup">Université de Lyon</orgName>
</affiliation>
</author>
</analytic>
<idno type="DOI">10.5220/0005297200840091</idno>
</biblStruct>
</sourceDesc>
</fileDesc>
<profileDesc><textClass><keywords scheme="mix" xml:lang="en"><term> Convolutional Neural Network</term>
<term> Multi-Layer Perceptron</term>
<term> OCR</term>
<term> Text Image</term>
<term>Super-Resolution</term>
</keywords>
</textClass>
</profileDesc>
</teiHeader>
<front><div type="abstract" xml:lang="en">We compare the performances of several Multi-Layer Perceptrons (MLPs) and Convolutional Neural Networks (ConvNets) for single text image Super-Resolution. We propose an example-based framework for both MLP and ConvNet, where a non-linear mapping between pairs of patches and high-frequency pixel values is learned. We then demonstrate that for equivalent complexity, ConvNets are better than MLPs at predicting missing details in upsampled text images. To evaluate the performances, we make use of a recent database (ULR-textSISR-2013a) along with different quality measures. We show that the proposed methods outperforms sparse coding-based methods for this database.</div>
</front>
</TEI>
<affiliations><list><country><li>France</li>
</country>
<region><li>Auvergne-Rhône-Alpes</li>
<li>Rhône-Alpes</li>
</region>
<settlement><li>Lyon</li>
</settlement>
<orgName><li>Université Claude Bernard Lyon 1</li>
<li>Université de Lyon</li>
</orgName>
</list>
<tree><country name="France"><noRegion><name sortKey="Peyrard, Clement" sort="Peyrard, Clement" uniqKey="Peyrard C" first="Clément" last="Peyrard">Clément Peyrard</name>
</noRegion>
<name sortKey="Garcia, Christophe" sort="Garcia, Christophe" uniqKey="Garcia C" first="Christophe" last="Garcia">Christophe Garcia</name>
<name sortKey="Mamalet, Franck" sort="Mamalet, Franck" uniqKey="Mamalet F" first="Franck" last="Mamalet">Franck Mamalet</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 000025 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Main/Exploration/biblio.hfd -nk 000025 | 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= Hal:hal-01260671 |texte= A Comparison between Multi-Layer Perceptrons and Convolutional Neural Networks for Text Image Super-Resolution }}
![]() | This area was generated with Dilib version V0.6.32. | ![]() |