Share Email Print
cover

Proceedings Paper

Harvesting weakly tagged images for computer vision tasks
Author(s): Yi Shen; Chunlei Yang; Yuli Gao; Jianping Fan
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

To crawl large amounts of weakly-tagged images for computer vision tasks such as object detection and scene recognition, it is very important to develop new techniques for tag cleansing and word sense disambiguation (i.e., removing irrelevant images from the crawled results). Based on this observation, a topic network is first generated to characterize both the semantic similarity contexts and the visual similarity contexts between the image topics more sufficiently. The topic network is used to represent the classes of objects and scenes of interest. Second, both the visual similarity contexts between the images and the semantic similarity contexts between their tags are integrated for tag cleansing and word sense disambiguation. By addressing the issues of polysemes and synonyms more effectively, our word sense disambiguation algorithm can determine the relevance between the images and the associated tags more precisely, and thus it can allow us to crawl large-scale weakly-tagged images for computer vision tasks.

Paper Details

Date Published: 10 February 2010
PDF: 9 pages
Proc. SPIE 7540, Imaging and Printing in a Web 2.0 World; and Multimedia Content Access: Algorithms and Systems IV, 75400Z (10 February 2010); doi: 10.1117/12.839291
Show Author Affiliations
Yi Shen, The Univ. of North Carolina at Charlotte (United States)
Chunlei Yang, The Univ. of North Carolina at Charlotte (United States)
Yuli Gao, Hewlett-Packard Labs. (United States)
Jianping Fan, The Univ. of North Carolina at Charlotte (United States)


Published in SPIE Proceedings Vol. 7540:
Imaging and Printing in a Web 2.0 World; and Multimedia Content Access: Algorithms and Systems IV
Theo Gevers; Qian Lin; Raimondo Schettini; Zhigang Fan; Cees Snoek, Editor(s)

© SPIE. Terms of Use
Back to Top