Publikation: SUR-Net : Predicting the Satisfied User Ratio Curve for Image Compression with Deep Learning
Dateien
Datum
Autor:innen
Herausgeber:innen
ISSN der Zeitschrift
Electronic ISSN
ISBN
Bibliografische Daten
Verlag
Schriftenreihe
Auflagebezeichnung
DOI (zitierfähiger Link)
Internationale Patentnummer
Angaben zur Forschungsförderung
Projekt
Open Access-Veröffentlichung
Core Facility der Universität Konstanz
Titel in einer weiteren Sprache
Publikationstyp
Publikationsstatus
Erschienen in
Zusammenfassung
The Satisfied User Ratio (SUR) curve for a lossy image compression scheme, e.g., JPEG, characterizes the probability distribution of the Just Noticeable Difference (JND) level, the smallest distortion level that can be perceived by a subject. We propose the first deep learning approach to predict such SUR curves. Instead of the direct approach of regressing the SUR curve itself for a given reference image, our model is trained on pairs of images, original and compressed. Relying on a Siamese Convolutional Neural Network (CNN), feature pooling, a fully connected regression-head, and transfer learning, we achieved a good prediction performance. Experiments on the MCL-JCI dataset showed a mean Bhattacharyya distance between the predicted and the original JND distributions of only 0.072.
Zusammenfassung in einer weiteren Sprache
Fachgebiet (DDC)
Schlagwörter
Konferenz
Rezension
Zitieren
ISO 690
FAN, Chunling, Hanhe LIN, Vlad HOSU, Yun ZHANG, Qingshan JIANG, Raouf HAMZAOUI, Dietmar SAUPE, 2019. SUR-Net : Predicting the Satisfied User Ratio Curve for Image Compression with Deep Learning. 2019 Eleventh International Conference on Quality of Multimedia Experience (QoMEX). Berlin, 5. Juni 2019 - 7. Juni 2019. In: IEEE COMPUTER SOCIETY, /, ed.. 2019 Eleventh International Conference on Quality of Multimedia Experience (QoMEX). Piscataway: IEEE, 2019, pp. 167-173. ISBN 978-1-5386-8212-8. Available under: doi: 10.1109/QoMEX.2019.8743204BibTex
@inproceedings{Fan2019SURNe-49118, year={2019}, doi={10.1109/QoMEX.2019.8743204}, title={SUR-Net : Predicting the Satisfied User Ratio Curve for Image Compression with Deep Learning}, isbn={978-1-5386-8212-8}, publisher={IEEE}, address={Piscataway}, booktitle={2019 Eleventh International Conference on Quality of Multimedia Experience (QoMEX)}, pages={167--173}, editor={IEEE Computer Society, /}, author={Fan, Chunling and Lin, Hanhe and Hosu, Vlad and Zhang, Yun and Jiang, Qingshan and Hamzaoui, Raouf and Saupe, Dietmar} }
RDF
<rdf:RDF xmlns:dcterms="http://purl.org/dc/terms/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:bibo="http://purl.org/ontology/bibo/" xmlns:dspace="http://digital-repositories.org/ontologies/dspace/0.1.0#" xmlns:foaf="http://xmlns.com/foaf/0.1/" xmlns:void="http://rdfs.org/ns/void#" xmlns:xsd="http://www.w3.org/2001/XMLSchema#" > <rdf:Description rdf:about="https://kops.uni-konstanz.de/server/rdf/resource/123456789/49118"> <void:sparqlEndpoint rdf:resource="http://localhost/fuseki/dspace/sparql"/> <dcterms:abstract xml:lang="eng">The Satisfied User Ratio (SUR) curve for a lossy image compression scheme, e.g., JPEG, characterizes the probability distribution of the Just Noticeable Difference (JND) level, the smallest distortion level that can be perceived by a subject. We propose the first deep learning approach to predict such SUR curves. Instead of the direct approach of regressing the SUR curve itself for a given reference image, our model is trained on pairs of images, original and compressed. Relying on a Siamese Convolutional Neural Network (CNN), feature pooling, a fully connected regression-head, and transfer learning, we achieved a good prediction performance. Experiments on the MCL-JCI dataset showed a mean Bhattacharyya distance between the predicted and the original JND distributions of only 0.072.</dcterms:abstract> <dc:creator>Hosu, Vlad</dc:creator> <dcterms:isPartOf rdf:resource="https://kops.uni-konstanz.de/server/rdf/resource/123456789/36"/> <dcterms:title>SUR-Net : Predicting the Satisfied User Ratio Curve for Image Compression with Deep Learning</dcterms:title> <dc:contributor>Jiang, Qingshan</dc:contributor> <dc:contributor>Hamzaoui, Raouf</dc:contributor> <dc:contributor>Zhang, Yun</dc:contributor> <dc:contributor>Fan, Chunling</dc:contributor> <dcterms:available rdf:datatype="http://www.w3.org/2001/XMLSchema#dateTime">2020-03-23T09:05:30Z</dcterms:available> <dcterms:issued>2019</dcterms:issued> <dc:contributor>Hosu, Vlad</dc:contributor> <dspace:isPartOfCollection rdf:resource="https://kops.uni-konstanz.de/server/rdf/resource/123456789/36"/> <dc:contributor>Saupe, Dietmar</dc:contributor> <dc:language>eng</dc:language> <dc:contributor>Lin, Hanhe</dc:contributor> <foaf:homepage rdf:resource="http://localhost:8080/"/> <dc:creator>Fan, Chunling</dc:creator> <dc:creator>Jiang, Qingshan</dc:creator> <bibo:uri rdf:resource="https://kops.uni-konstanz.de/handle/123456789/49118"/> <dc:creator>Zhang, Yun</dc:creator> <dc:creator>Saupe, Dietmar</dc:creator> <dc:creator>Hamzaoui, Raouf</dc:creator> <dc:date rdf:datatype="http://www.w3.org/2001/XMLSchema#dateTime">2020-03-23T09:05:30Z</dc:date> <dc:creator>Lin, Hanhe</dc:creator> </rdf:Description> </rdf:RDF>