Department of Computer Science
Collaborative and content-based image labeling
Many on-line photo sharing systems allow users to tag their images so as to support semantic image search. In this paper, we study how one can take advantages of the already-tagged images to (semi-)automate the labeling of newly uploaded ones. In particular, we propose a hybrid approach for the prediction where user-provided tags and image visual contents are fused under a unified probabilistic framework. Kernel smoothing and collaborative filtering techniques are explored for improving the accuracy of the probabilistic models estimation. By comparing with some state-of-the-art content-based image labeling methods, we have empirically shown that 1) the proposed method can achieve comparable tag prediction accuracy when there is no user-provided tag, and that 2) it can significantly boost the prediction accuracy if the user can provide just a few tags.
Collaboration, Labeling, Filtering, Image retrieval, Computer science, Accuracy, Kernel, Smoothing methods, Content based retrieval, Tagging
Source Publication Title
19th International Conference on Pattern Recognition, 2008 ICPR 2008 ; 8 - 11 Dec. 2008, Tampa, Florida, USA
International Association for Pattern Recognition
Tampa, United States
This work was supported in part by MoE research Fund under contract 104075, Shanghai Municipal R&D Foundation under contract 06DZ15008, and MoST Support Program under contract 2007BAH09B03.
Link to Publisher's Edition
Zhou, Ning, William K. Cheung, Xiangyang Xue, and Guoping Qiu. "Collaborative and content-based image labeling." 19th International Conference on Pattern Recognition, 2008 ICPR 2008 ; 8 - 11 Dec. 2008, Tampa, Florida, USA (2008).