Domain adaptation on the statistical manifold

Baktashmotlagh, Mahsa, Harandi, Mehrtash T., Lovell, Brian C. and Salzmann, Mathieu (2014). Domain adaptation on the statistical manifold. In: Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. 27th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2014, Columbus, OH, United States, (2481-2488). 23-28 June 2014. doi:10.1109/CVPR.2014.318


Author Baktashmotlagh, Mahsa
Harandi, Mehrtash T.
Lovell, Brian C.
Salzmann, Mathieu
Title of paper Domain adaptation on the statistical manifold
Conference name 27th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2014
Conference location Columbus, OH, United States
Conference dates 23-28 June 2014
Convener IEEE
Proceedings title Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition   Check publisher's open access policy
Journal name IEEE Conference on Computer Vision and Pattern Recognition. Proceedings   Check publisher's open access policy
Place of Publication Piscataway, NJ, United States
Publisher I E E E Computer Society
Publication Year 2014
Year available 2014
Sub-type Fully published paper
DOI 10.1109/CVPR.2014.318
Open Access Status Not Open Access
ISBN 9781479951178
ISSN 1063-6919
Start page 2481
End page 2488
Total pages 8
Language eng
Abstract/Summary In this paper, we tackle the problem of unsupervised domain adaptation for classification. In the unsupervised scenario where no labeled samples from the target domain are provided, a popular approach consists in transforming the data such that the source and target distributions become similar. To compare the two distributions, existing approaches make use of the Maximum Mean Discrepancy (MMD). However, this does not exploit the fact that probability distributions lie on a Riemannian manifold. Here, we propose to make better use of the structure of this manifold and rely on the distance on the manifold to compare the source and target distributions. In this framework, we introduce a sample selection method and a subspace-based method for unsupervised domain adaptation, and show that both these manifold-based techniques outperform the corresponding approaches based on the MMD. Furthermore, we show that our subspace-based approach yields state-of-the-art results on a standard object recognition benchmark.
Keyword Manifolds
Measurement
Kernel
Optimization
Geometry
Object recognition
Visualization
Q-Index Code C1
Q-Index Status Provisional Code
Institutional Status UQ

 
Versions
Version Filter Type
Citation counts: TR Web of Science Citation Count  Cited 4 times in Thomson Reuters Web of Science Article | Citations
Scopus Citation Count Cited 4 times in Scopus Article | Citations
Google Scholar Search Google Scholar
Created: Sun, 25 Oct 2015, 00:26:53 EST by System User on behalf of Learning and Research Services (UQ Library)