Unsupervised domain adaptation by Domain Invariant Projection

Baktashmotlagh, Mahsa, Harandi, Mehrtash T., Lovell, Brian C. and Salzmann, Mathieu (2013). Unsupervised domain adaptation by Domain Invariant Projection. In: Proceedings: 2013 IEEE International Conference on Computer Vision. 2013 IEEE International Conference on Computer Vision (ICCV), Sydney, Australia, (769-776). 1-8 December 2013. doi:10.1109/ICCV.2013.100

Attached Files (Some files may be inaccessible until you login with your UQ eSpace credentials)
Name Description MIMEType Size Downloads

Author Baktashmotlagh, Mahsa
Harandi, Mehrtash T.
Lovell, Brian C.
Salzmann, Mathieu
Title of paper Unsupervised domain adaptation by Domain Invariant Projection
Conference name 2013 IEEE International Conference on Computer Vision (ICCV)
Conference location Sydney, Australia
Conference dates 1-8 December 2013
Proceedings title Proceedings: 2013 IEEE International Conference on Computer Vision   Check publisher's open access policy
Journal name IEEE International Conference on Computer Vision. Proceedings   Check publisher's open access policy
Place of Publication Piscataway, NJ, United States
Publisher IEEE
Publication Year 2013
Sub-type Fully published paper
DOI 10.1109/ICCV.2013.100
Open Access Status
ISBN 9781479928392
ISSN 1550-5499
Start page 769
End page 776
Total pages 8
Language eng
Abstract/Summary Domain-invariant representations are key to addressing the domain shift problem where the training and test examples follow different distributions. Existing techniques that have attempted to match the distributions of the source and target domains typically compare these distributions in the original feature space. This space, however, may not be directly suitable for such a comparison, since some of the features may have been distorted by the domain shift, or may be domain specific. In this paper, we introduce a Domain Invariant Projection approach: An unsupervised domain adaptation method that overcomes this issue by extracting the information that is invariant across the source and target domains. More specifically, we learn a projection of the data to a low-dimensional latent space where the distance between the empirical distributions of the source and target examples is minimized. We demonstrate the effectiveness of our approach on the task of visual object recognition and show that it outperforms state-of-the-art methods on a standard domain adaptation benchmark dataset.
Q-Index Code EX
Q-Index Status Confirmed Code
Institutional Status UQ

 
Versions
Version Filter Type
Citation counts: TR Web of Science Citation Count  Cited 27 times in Thomson Reuters Web of Science Article | Citations
Scopus Citation Count Cited 62 times in Scopus Article | Citations
Google Scholar Search Google Scholar
Created: Tue, 08 Apr 2014, 04:13:51 EST by Ms Deborah Brian on behalf of School of Information Technol and Elec Engineering