Robust semantic video indexing by harvesting web images

Yang, Yang, Zha, Zheng-Jun, Shen, Heng Tao and Chua, Tat-Seng (2013). Robust semantic video indexing by harvesting web images. In: Shipeng Li, Abdulmotaleb El Saddik, Meng Wang, Tao Mei, Nicu Sebe, Shuicheng Yan, Richang Hong and Cathal Gurrin, Advances in Multimedia Modeling - 19th International Conference, MMM 2013, Proceedings. 19th International Conference on Multimedia Modeling (MMM 2013), Huangshan, China, (70-80). 7 - 9 January 2013. doi:10.1007/978-3-642-35725-1_7

Author Yang, Yang
Zha, Zheng-Jun
Shen, Heng Tao
Chua, Tat-Seng
Title of paper Robust semantic video indexing by harvesting web images
Conference name 19th International Conference on Multimedia Modeling (MMM 2013)
Conference location Huangshan, China
Conference dates 7 - 9 January 2013
Proceedings title Advances in Multimedia Modeling - 19th International Conference, MMM 2013, Proceedings   Check publisher's open access policy
Journal name Lecture Notes in Computer Science   Check publisher's open access policy
Place of Publication Heidelberg, Germany
Publisher Springer
Publication Year 2013
Year available 2013
Sub-type Fully published paper
DOI 10.1007/978-3-642-35725-1_7
ISBN 9783642357244
ISSN 0302-9743
Editor Shipeng Li
Abdulmotaleb El Saddik
Meng Wang
Tao Mei
Nicu Sebe
Shuicheng Yan
Richang Hong
Cathal Gurrin
Volume 7732
Issue PART 1
Start page 70
End page 80
Total pages 11
Collection year 2014
Language eng
Abstract/Summary Semantic video indexing, also known as video annotation, video concept detection in literatures, has attracted significant attentions recently. Due to the scarcity of training videos, most existing approaches can scarcely achieve satisfactory performances. This paper proposes a robust semantic video indexing framework, which exploits user-tagged web images to assist learning robust semantic video indexing classifiers. The following two challenges are well studied: (a) domain difference between images and videos; and (b) noisy web images with incorrect tags. Specifically, we first estimate the probabilities of images being correctly tagged as confidence scores and filter out the images with low confidence scores. We then develop a robust image-to-video indexing approach to learn reliable classifiers from a limited number of training videos together with abundant user-tagged images. A robust loss function weighted by the confidence scores of images is used to further alleviate the influence of noisy samples. An optimal kernel space, in which the domain difference between images and videos is minimal, is automatically discovered by the approach to tackle the domain difference problem. Experiments on NUS-WIDE web image dataset and Kodak consumer video corpus demonstrate the effectiveness of the proposed robust semantic video indexing framework.
Subjects 1700 Computer Science
2614 Theoretical Computer Science
Keyword Robust semantic video
Semantic video indexing
Video concept detection in literature
Q-Index Code C1
Q-Index Status Confirmed Code
Institutional Status UQ

Version Filter Type
Citation counts: Scopus Citation Count Cited 2 times in Scopus Article | Citations
Google Scholar Search Google Scholar
Created: Tue, 18 Feb 2014, 11:26:12 EST by System User on behalf of School of Information Technol and Elec Engineering