Type of Publication: | Contribution to a conference collection |
Publication status: | Published |
Author: | Lan Ha, Mai; Hosu, Vlad; Blanz, Volker |
Year of publication: | 2020 |
Conference: | 2020 IEEE Winter Conference on Applications of Computer Vision (WACV), Mar 1, 2020 - Mar 5, 2020, Snowmass, CO |
Published in: | 2020 IEEE Winter Conference on Applications of Computer Vision (WACV). - Piscataway, NJ : IEEE, 2020. - pp. 2548-2557. - eISSN 2642-9381. - ISBN 978-1-72816-553-0 |
DOI (citable link): | https://dx.doi.org/10.1109/WACV45572.2020.9093522 |
Summary: |
Assessing visual similarity in-the-wild, a core ability of the human visual system, is a challenging problem for computer vision methods because of its subjective nature and limited annotated datasets. We make a stride forward, showing that visual similarity can be better studied by isolating its components. We identify color composition similarity as an important aspect and study its interaction with category-level similarity. Color composition similarity considers the distribution of colors and their layout in images. We create predictive models accounting for the global similarity that is beyond pixel-based and patch-based, or histogram level information. Using an active learning approach, we build a large-scale color composition similarity dataset with subjective ratings via crowd-sourcing, the first of its kind. We train a Siamese network using the dataset to create a color similarity metric and descriptors which outperform existing color descriptors. We also provide a benchmark for global color descriptors for perceptual color similarity. Finally, we combine color similarity and category level features for fine-grained visual similarity. Our proposed model surpasses the state-of-the-art performance while using three orders of magnitude less training data. The results suggest that our proposal to study visual similarity by isolating its components, modeling and combining them is a promising paradigm for further development.
|
Subject (DDC): | 004 Computer Science |
Bibliography of Konstanz: | Yes |
Files | Size | Format | View |
---|---|---|---|
There are no files associated with this item. |
LAN HA, Mai, Vlad HOSU, Volker BLANZ, 2020. Color Composition Similarity and Its Application in Fine-grained Similarity. 2020 IEEE Winter Conference on Applications of Computer Vision (WACV). Snowmass, CO, Mar 1, 2020 - Mar 5, 2020. In: 2020 IEEE Winter Conference on Applications of Computer Vision (WACV). Piscataway, NJ:IEEE, pp. 2548-2557. eISSN 2642-9381. ISBN 978-1-72816-553-0. Available under: doi: 10.1109/WACV45572.2020.9093522
@inproceedings{LanHa2020Color-53099, title={Color Composition Similarity and Its Application in Fine-grained Similarity}, year={2020}, doi={10.1109/WACV45572.2020.9093522}, isbn={978-1-72816-553-0}, address={Piscataway, NJ}, publisher={IEEE}, booktitle={2020 IEEE Winter Conference on Applications of Computer Vision (WACV)}, pages={2548--2557}, author={Lan Ha, Mai and Hosu, Vlad and Blanz, Volker} }
<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/123456789/53099"> <dc:language>eng</dc:language> <void:sparqlEndpoint rdf:resource="http://localhost/fuseki/dspace/sparql"/> <dcterms:issued>2020</dcterms:issued> <dc:creator>Hosu, Vlad</dc:creator> <dcterms:title>Color Composition Similarity and Its Application in Fine-grained Similarity</dcterms:title> <dspace:isPartOfCollection rdf:resource="https://kops.uni-konstanz.de/rdf/resource/123456789/36"/> <foaf:homepage rdf:resource="http://localhost:8080/jspui"/> <dc:contributor>Blanz, Volker</dc:contributor> <dc:creator>Lan Ha, Mai</dc:creator> <bibo:uri rdf:resource="https://kops.uni-konstanz.de/handle/123456789/53099"/> <dc:date rdf:datatype="http://www.w3.org/2001/XMLSchema#dateTime">2021-03-08T10:04:07Z</dc:date> <dcterms:isPartOf rdf:resource="https://kops.uni-konstanz.de/rdf/resource/123456789/36"/> <dc:creator>Blanz, Volker</dc:creator> <dcterms:abstract xml:lang="eng">Assessing visual similarity in-the-wild, a core ability of the human visual system, is a challenging problem for computer vision methods because of its subjective nature and limited annotated datasets. We make a stride forward, showing that visual similarity can be better studied by isolating its components. We identify color composition similarity as an important aspect and study its interaction with category-level similarity. Color composition similarity considers the distribution of colors and their layout in images. We create predictive models accounting for the global similarity that is beyond pixel-based and patch-based, or histogram level information. Using an active learning approach, we build a large-scale color composition similarity dataset with subjective ratings via crowd-sourcing, the first of its kind. We train a Siamese network using the dataset to create a color similarity metric and descriptors which outperform existing color descriptors. We also provide a benchmark for global color descriptors for perceptual color similarity. Finally, we combine color similarity and category level features for fine-grained visual similarity. Our proposed model surpasses the state-of-the-art performance while using three orders of magnitude less training data. The results suggest that our proposal to study visual similarity by isolating its components, modeling and combining them is a promising paradigm for further development.</dcterms:abstract> <dc:contributor>Lan Ha, Mai</dc:contributor> <dc:contributor>Hosu, Vlad</dc:contributor> <dcterms:available rdf:datatype="http://www.w3.org/2001/XMLSchema#dateTime">2021-03-08T10:04:07Z</dcterms:available> </rdf:Description> </rdf:RDF>