Visualization and Intelligent Systems Laboratory
VISLab

 

 

Contact Information

VISLab
Winston Chung Hall Room 216
University of California, Riverside
900 University Avenue
Riverside, CA 92521-0425


Tel: (951)-827-3954

CRIS
Bourns College of Engineering
UCR
NSF IGERT on Video Bioinformatics

UCR Collaborators:
CSE
ECE
ME
STAT
PSYC
ENTM
BIOL
BPSC
ECON
MATH
BIOENG
MGNT

Other Collaborators:
Keio University

Other Activities:
IEEE Biometrics Workshop 2014
IEEE Biometrics Workshop 2013
Worshop on DVSN 2009
Multibiometrics Book

Webmaster Contact Information:
Alex Shin
wshin@ece.ucr.edu

Last updated: July 1, 2017

 

 

On Mixture Model for Image Databases

Presented by: Anlei Dong

Abstract:

Recently mixture model has been used to model image databases. The retrieval experiences derived from multiple users' relevance feedbacks have been exploited to improve model fitting in a semi-supervised manner. However, the mixture model for image databases remains a challenging task since the database may contain clutter (the images not belonging to any class) and outliers (the images being far away from the components corresponding to the classes they belong to), and labeling information derived from multiple users may be inconsistent. Thus, neither the mixture model nor the labeling information is as ideal as most of the researchers have previously assumed. In this paper, we (a) address the challenges to handle the noise disturbances for both mixture model and users' labeling information, (b) propose to process retrieval experiences in an intelligent manner using Bayesian analysis, (c) provide a novel probabilistic multiple discriminant analysis for feature dimensionality reduction, (d) present a robust mixture model fitting algorithm to achieve visual concept learning, and (e) construct a concept-based indexing structure for efficient search of the database. The experimental results on two image databases show the correctness of our retrieval experience analysis, the effectiveness of the proposed concept learning approach, and the improvement of retrieval performance based on the indexing structure.