Personalized image retrieval in compressed domain based on user interest model

Authors:
Zhenwei Li;Jing Zhang;Li Zhuo;Mengmeng Diao
Affiliations:
Beijing University of Technology, Beijing, P.R. China;Beijing University of Technology, Beijing, P.R. China;Beijing University of Technology, Beijing, P.R. China;Beijing University of Technology, Beijing, P.R. China
Venue:
Proceedings of the Fifth International Conference on Internet Multimedia Computing and Service
Year:
2013

Citing 6
Cited 0

Unified video annotation via multigraph learning

IEEE Transactions on Circuits and Systems for Video Technology
The Quantitative Model of User Interests Based on Web Log

ICCSEE '12 Proceedings of the 2012 International Conference on Computer Science and Electronics Engineering - Volume 03
Towards a Relevant and Diverse Search of Social Images

IEEE Transactions on Multimedia
Compressed domain content based retrieval using H.264 DC-pictures

Multimedia Tools and Applications
Query specific fusion for image retrieval

ECCV'12 Proceedings of the 12th European conference on Computer Vision - Volume Part II
An approach of bag-of-words based on visual attention model for pornographic images recognition in compressed domain

Neurocomputing

Quantified Score

Hi-index	0.00

Visualization

Abstract

With the rapid development of image retrieval technology, personalized image retrieval has attracted widespread attention. On the Internet, a great deal of images are stored and transmitted in a compressed format. In order to improve the accuracy and reduce the decoding time in the process of image retrieval, personalized image retrieval in compressed domain based on user interest model is proposed in this paper. First, according to the JPEG compressed format, the low resolution image is constructed to extract its visual features. Second, the user interest model is utilized to realize personalized image retrieval. At last, the user interest model is updated with user relevant feedback of short-term interest and long-term interest. Experimental results show that the proposed method can significantly reduce the time of image retrieval, as well as improving recall and precision.