Optimal graph leaning with partial tags and multiple features for image and video annotation

Gao, Lianli, Song, Jingkuan, Nie, Feiping, Yan, Yan, Sebe, Nicu and Shen, Heng Tao (2015). Optimal graph leaning with partial tags and multiple features for image and video annotation. In: 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). IEEE Conference on Computer Vision and Pattern Recognition, Boston, MA, United States, (4371-4379). 7-12 June 2015. doi:10.1109/CVPR.2015.7299066

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

Author Gao, Lianli
Song, Jingkuan
Nie, Feiping
Yan, Yan
Sebe, Nicu
Shen, Heng Tao
Title of paper Optimal graph leaning with partial tags and multiple features for image and video annotation
Conference name IEEE Conference on Computer Vision and Pattern Recognition
Conference location Boston, MA, United States
Conference dates 7-12 June 2015
Convener IEEE
Proceedings title 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)   Check publisher's open access policy
Journal name Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition   Check publisher's open access policy
Place of Publication Piscataway, NJ, United States
Publisher IEEE
Publication Year 2015
Sub-type Fully published paper
DOI 10.1109/CVPR.2015.7299066
Open Access Status Not Open Access
ISBN 9781467369640
ISSN 1063-6919
Volume 07-12-June-2015
Start page 4371
End page 4379
Total pages 9
Collection year 2016
Language eng
Formatted Abstract/Summary
In multimedia annotation, due to the time constraints and the tediousness of manual tagging, it is quite common to utilize both tagged and untagged data to improve the performance of supervised learning when only limited tagged training data are available. This is often done by adding a geometrically based regularization term in the objective function of a supervised learning model. In this case, a similarity graph is indispensable to exploit the geometrical relationships among the training data points, and the graph construction scheme essentially determines the performance of these graph-based learning algorithms. However, most of the existing works construct the graph empirically and are usually based on a single feature without using the label information. In this paper, we propose a semi-supervised annotation approach by learning an optimal graph (OGL) from multi-cues (i.e., partial tags and multiple features) which can more accurately embed the relationships among the data points. We further extend our model to address out-of-sample and noisy label issues. Extensive experiments on four public datasets show the consistent superiority of OGL over state-of-the-art methods by up to 12% in terms of mean average precision.
Q-Index Code E1
Q-Index Status Confirmed Code
Institutional Status UQ

 
Versions
Version Filter Type
Citation counts: Scopus Citation Count Cited 9 times in Scopus Article | Citations
Google Scholar Search Google Scholar
Created: Mon, 09 Mar 2015, 15:33:35 EST by Dr Heng Tao Shen on behalf of School of Information Technol and Elec Engineering