Technical Program

Paper Detail

Paper:MMSA-P2.4
Session:Intelligent Processing Techniques for Multimedia Systems and Applications
Time:Monday, May 24, 09:30 - 11:00
Presentation: Poster
Topic: Multimedia Systems and Applications: Multimedia Database and Retrieval Systems
Title: PROBABILISTIC APPROACH TO K-NEAREST NEIGHBOR VIDEO RETRIEVAL
Authors: Nai-xiang Lian; Nanyang Technological University 
 Yap-Peng Tan; Nanyang Technological University 
Abstract: In this paper we propose a probabilistic approach to retrieve video clips similar to a given query video clip. In our approach the video clips are partitioned into video segments based on their content homogeneity, and video segments indatabase are connected during the query process to construct candidate clips and compare with the query clip for theirsimilarity(or distance). An efficient scheme is developed to estimate the probability density functions of distancesbetween candidate clips and a query clip, and based on these density functions, two methods are devised to reduce thenumber of candidate clips to be compared for speeding up the retrieval process. Experimental results show that our proposed approach can notably speed up the retrieval of similar video clip, while maintaining high retrieval accuracy.
 
           Back


Home -||- Technical Program -||- Call for Participation -||- Paper Submission -||- Paper Review -||- Plenaries
Invited Sessions -||- Registration -||- Tutorials/Short Courses -||- Forums -||- Technical Program Committee
MySchedule -||- Exhibits -||- Housing -||- Information -||- Committee -||- Contact Us

©2021 Conference Management Services -||- email: organizers@iscas2004.org -||- Last updated Friday, August 17, 2012