skip to main content
10.1145/1180639.1180667acmconferencesArticle/Chapter ViewAbstractPublication PagesmmConference Proceedingsconference-collections
Article

To construct optimal training set for video annotation

Published: 23 October 2006 Publication History

Abstract

This paper exploits the criteria to optimize the training set construction for video annotation. Most existing learning-based semantic annotation approaches require a large training set to achieve good generalization capacity, in which a considerable amount of labor-intensively manual labeling is desirable. However, it is observed that the generalization capacity of a classifier highly depends on the geometrical distribution rather than the size of the training data. We argue that a training set which includes most temporal and spatial distribution of the whole data will achieve a satisfying performance even in the case of limited size of training set. In order to capture the geometrical distribution characteristics of a given video collection, we propose the following four metrics for constructing an optimal training set, including Salience Time Dispersiveness Spatial Dispersiveness and Diversity. Moreover, based on these metrics, we propose a set of optimization rules to capture the most distribution information of the whole data for a training set with a given size. Experimental results demonstrate that these rules are effective for training set construction for video annotation, and significantly outperform random training set selection as well.

References

[1]
Guidelines for the trecvid 2003 evaluation. http://wwwnlpir.nist.gov/projects/tv2003/tv2003.htm.
[2]
G. Baudat and F. Anouar. Feature vector selection and projection using kernels. Neurocomputing, 55(9):21--38, 2003.
[3]
J. Fan, H. Luo, and X. Lin. Semantic video classification by integrating flexible mixture model with adaptive em algorithm. In Proceedings of ACM SIGMM International Workshop on Multimedia Information Retrieval, pages 9--16, Nov 2003.
[4]
K.-S. Goh, E. Chang, and W.-C. Lai. Concept-dependent multimodal active learning for image retrieval. In Proceedings of ACM International Conference on Multimedia, pages 564--571, Oct 2004.
[5]
G.-J. Qi, Y. Song, X.-S. Hua, H.-J. Zhang, and D.-R. Dai. Video annotation by active learning and cluster tuning. In Proceedings of International Workshop on Semantic Learning Applications in Multimedia, June 2006.
[6]
Y. Song, X.-S. Hua, L.-R. Dai, and M. Wang. Semi-automatic video annotation based on active learning with multiple complementary predictors. In Proceedings of ACM SIGMM International Workshop on Multimedia Information Retrieval, pages 97--104, Nov 2005.
[7]
V. Vapnik. Three remarks on support vector method of function estimation. Advances in Kernel Methods: Support Vector Learning. MIT Press, 1999.
[8]
M. Wang, X.-S. Hua, L.-R.Dai, and Y.Song.Enhanced semi-supervised learning for automatic video annotation. In Proceedings of IEEE International Conference in Multimedia & Expo, pages 1485--1488, July 2006.
[9]
L. Xie, S.-F. Chang, A. Divakaran, and H. Sun. Structure analysis of soccer video with hidden markov models. In Proceedings of IEEE International Conference on Acoustics, Speech and Signal Processing, pages 13--17, May 2002.
[10]
R. Yan and M. Naphade. Semi-supervised cross feature learning for semantic concept detection in videos. In Proceedings of IEEE International Conference on Computer Vision and Pattern Recognition, pages 657--663, June 2005.
[11]
R. Yan, J. Yang, and A. Hauptmann. Automatically labeling video data using multi-class active learning. In Proceedings of IEEE International Conference on Computer Vision,pages 516--523, Oct 2003.
[12]
D. Zhong and S.-F. Chang. Structure analysis of sports video using domain models. In Proceedings of IEEE International Conference in Multimedia & Expo, pages 713--716, Aug 2001.

Cited By

View all
  • (2018)Typicality rankingMultimedia Tools and Applications10.1007/s11042-011-0892-070:2(647-660)Online publication date: 31-Dec-2018
  • (2018)Facebook5k: A Novel Evaluation Resource Dataset for Cross-Media SearchCloud Computing and Security10.1007/978-3-030-00006-6_47(512-524)Online publication date: 1-Nov-2018
  • (2011)A novel method for semantic video concept learning using web imagesProceedings of the 19th ACM international conference on Multimedia10.1145/2072298.2071943(1081-1084)Online publication date: 28-Nov-2011
  • Show More Cited By

Recommendations

Comments

Information & Contributors

Information

Published In

cover image ACM Conferences
MM '06: Proceedings of the 14th ACM international conference on Multimedia
October 2006
1072 pages
ISBN:1595934472
DOI:10.1145/1180639
Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from [email protected]

Sponsors

Publisher

Association for Computing Machinery

New York, NY, United States

Publication History

Published: 23 October 2006

Permissions

Request permissions for this article.

Check for updates

Author Tags

  1. training set construction
  2. video annotation

Qualifiers

  • Article

Conference

MM06
MM06: The 14th ACM International Conference on Multimedia 2006
October 23 - 27, 2006
CA, Santa Barbara, USA

Acceptance Rates

Overall Acceptance Rate 2,145 of 8,556 submissions, 25%

Contributors

Other Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

  • Downloads (Last 12 months)3
  • Downloads (Last 6 weeks)1
Reflects downloads up to 16 Feb 2025

Other Metrics

Citations

Cited By

View all
  • (2018)Typicality rankingMultimedia Tools and Applications10.1007/s11042-011-0892-070:2(647-660)Online publication date: 31-Dec-2018
  • (2018)Facebook5k: A Novel Evaluation Resource Dataset for Cross-Media SearchCloud Computing and Security10.1007/978-3-030-00006-6_47(512-524)Online publication date: 1-Nov-2018
  • (2011)A novel method for semantic video concept learning using web imagesProceedings of the 19th ACM international conference on Multimedia10.1145/2072298.2071943(1081-1084)Online publication date: 28-Nov-2011
  • (2011)Determination of emotional content of video clips by low-level audiovisual featuresMultimedia Tools and Applications10.1007/s11042-010-0702-061:1(21-49)Online publication date: 11-Jan-2011
  • (2009)NUS-WIDEProceedings of the ACM International Conference on Image and Video Retrieval10.1145/1646396.1646452(1-9)Online publication date: 8-Jul-2009
  • (2008)Online multi-label active annotationProceedings of the 16th ACM international conference on Multimedia10.1145/1459359.1459379(141-150)Online publication date: 26-Oct-2008
  • (2007)Beyond Accuracy: Typicality Ranking for Video AnnotationMultimedia and Expo, 2007 IEEE International Conference on10.1109/ICME.2007.4284733(647-650)Online publication date: Jul-2007

View Options

Login options

View options

PDF

View or Download as a PDF file.

PDF

eReader

View online with eReader.

eReader

Figures

Tables

Media

Share

Share

Share this Publication link

Share on social media