Transfer deep convolutional activation-based features for domain adaptation in sensor networks
Transfer deep convolutional activation-based features for domain adaptation in sensor networks
About this item
Full title
Author / Creator
Publisher
Cham: Springer International Publishing
Journal title
Language
English
Formats
Publication information
Publisher
Cham: Springer International Publishing
Subjects
More information
Scope and Contents
Contents
In this paper, we propose a novel method named transfer deep convolutional activation-based features (TDCAF) for domain adaptation in sensor networks. Specifically, we first train a siamese network with weight sharing to map the images from different domains into the same feature space, which can learn domain-invariant information. Since various fe...
Alternative Titles
Full title
Transfer deep convolutional activation-based features for domain adaptation in sensor networks
Authors, Artists and Contributors
Author / Creator
Identifiers
Primary Identifiers
Record Identifier
TN_cdi_doaj_primary_oai_doaj_org_article_04a43cbbd9ee4fa7af5a2c416832f7d9
Permalink
https://devfeature-collection.sl.nsw.gov.au/record/TN_cdi_doaj_primary_oai_doaj_org_article_04a43cbbd9ee4fa7af5a2c416832f7d9
Other Identifiers
ISSN
1687-1499
E-ISSN
1687-1499
DOI
10.1186/s13638-018-1059-8
