Aufgrund von Vorbereitungen auf eine neue Version von KOPS, können am Montag, 6.2. und Dienstag, 7.2. keine Publikationen eingereicht werden. (Due to preparations for a new version of KOPS, no publications can be submitted on Monday, Feb. 6 and Tuesday, Feb. 7.)
Type of Publication: | Preprint |
Publication status: | Published |
URI (citable link): | http://nbn-resolving.de/urn:nbn:de:bsz:352-2-amv1uzjjhj4x0 |
Author: | Hosu, Vlad; Goldlücke, Bastian; Saupe, Dietmar |
Year of publication: | 2019 |
ArXiv-ID: | arXiv:1904.01382 |
Summary: |
We propose an effective deep learning approach to aesthetics quality assessment that relies on a new type of pre-trained features, and apply it to the AVA data set, the currently largest aesthetics database. While previous approaches miss some of the information in the original images, due to taking small crops, down-scaling or warping the originals during training, we propose the first method that efficiently supports full resolution images as an input, and can be trained on variable input sizes. This allows us to significantly improve upon the state of the art, increasing the Spearman rank-order correlation coefficient (SRCC) of ground-truth mean opinion scores (MOS) from the existing best reported of 0.612 to 0.756. To achieve this performance, we extract multi-level spatially pooled (MLSP) features from all convolutional blocks of a pre-trained InceptionResNet-v2 network, and train a custom shallow Convolutional Neural Network (CNN) architecture on these new features.
|
Subject (DDC): | 004 Computer Science |
Comment on publication: | To appear in CVPR 2019 |
Link to License: | In Copyright |
Bibliography of Konstanz: | Yes |
HOSU, Vlad, Bastian GOLDLÜCKE, Dietmar SAUPE, 2019. Effective Aesthetics Prediction with Multi-level Spatially Pooled Features
@unpublished{Hosu2019-04-02T12:58:12ZEffec-45609, title={Effective Aesthetics Prediction with Multi-level Spatially Pooled Features}, year={2019}, author={Hosu, Vlad and Goldlücke, Bastian and Saupe, Dietmar}, note={To appear in CVPR 2019} }
<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/rdf/resource/123456789/45609"> <dcterms:abstract xml:lang="eng">We propose an effective deep learning approach to aesthetics quality assessment that relies on a new type of pre-trained features, and apply it to the AVA data set, the currently largest aesthetics database. While previous approaches miss some of the information in the original images, due to taking small crops, down-scaling or warping the originals during training, we propose the first method that efficiently supports full resolution images as an input, and can be trained on variable input sizes. This allows us to significantly improve upon the state of the art, increasing the Spearman rank-order correlation coefficient (SRCC) of ground-truth mean opinion scores (MOS) from the existing best reported of 0.612 to 0.756. To achieve this performance, we extract multi-level spatially pooled (MLSP) features from all convolutional blocks of a pre-trained InceptionResNet-v2 network, and train a custom shallow Convolutional Neural Network (CNN) architecture on these new features.</dcterms:abstract> <dc:creator>Goldlücke, Bastian</dc:creator> <dc:language>eng</dc:language> <bibo:uri rdf:resource="https://kops.uni-konstanz.de/handle/123456789/45609"/> <dcterms:available rdf:datatype="http://www.w3.org/2001/XMLSchema#dateTime">2019-04-09T13:48:49Z</dcterms:available> <dcterms:rights rdf:resource="https://rightsstatements.org/page/InC/1.0/"/> <dcterms:issued>2019-04-02T12:58:12Z</dcterms:issued> <dc:contributor>Saupe, Dietmar</dc:contributor> <dspace:isPartOfCollection rdf:resource="https://kops.uni-konstanz.de/rdf/resource/123456789/36"/> <dcterms:title>Effective Aesthetics Prediction with Multi-level Spatially Pooled Features</dcterms:title> <foaf:homepage rdf:resource="http://localhost:8080/jspui"/> <dc:creator>Saupe, Dietmar</dc:creator> <void:sparqlEndpoint rdf:resource="http://localhost/fuseki/dspace/sparql"/> <dcterms:hasPart rdf:resource="https://kops.uni-konstanz.de/bitstream/123456789/45609/1/Hosu_2-amv1uzjjhj4x0.pdf"/> <dcterms:isPartOf rdf:resource="https://kops.uni-konstanz.de/rdf/resource/123456789/36"/> <dc:creator>Hosu, Vlad</dc:creator> <dc:date rdf:datatype="http://www.w3.org/2001/XMLSchema#dateTime">2019-04-09T13:48:49Z</dc:date> <dc:rights>terms-of-use</dc:rights> <dspace:hasBitstream rdf:resource="https://kops.uni-konstanz.de/bitstream/123456789/45609/1/Hosu_2-amv1uzjjhj4x0.pdf"/> <dc:contributor>Hosu, Vlad</dc:contributor> <dc:contributor>Goldlücke, Bastian</dc:contributor> </rdf:Description> </rdf:RDF>
Hosu_2-amv1uzjjhj4x0.pdf | 185 |