Distributional distances in color image retrieval with GMVQ-Generated histograms

Authors:
Sangoh Jeong
Affiliations:
Dept. of Electrical Engineering, Stanford University, Stanford, CA
Venue:
CIVR'05 Proceedings of the 4th international conference on Image and Video Retrieval
Year:
2005

Citing 8
Cited 0

Vector quantization and signal compression

Vector quantization and signal compression
Color indexing

International Journal of Computer Vision
Regions-of-Interest and Spatial Layout for Content-Based Image Retrieval

Multimedia Tools and Applications
Retrieval Performance Improvement through Low Rank Corrections

CBAIVL '99 Proceedings of the IEEE Workshop on Content-Based Access of Image and Video Libraries
Image Segmentation by Jensen-Shannon Divergence. Application to Measurement of Interfacial Tension

ICPR '00 Proceedings of the International Conference on Pattern Recognition - Volume 3
Bayesian models for visual information retrieval

Bayesian models for visual information retrieval
Image retrieval using color histograms generated by Gauss mixture vector quantization

Computer Vision and Image Understanding - Special issue on color for image indexing and retrieval
Minimum Distortion Color Image Retrieval Based on Lloyd-Clustered Gauss Mixtures

DCC '05 Proceedings of the Data Compression Conference

Quantified Score

Hi-index	0.01

Visualization

Abstract

We investigate and compare the performance of several distributional distances in generic color image retrieval with an emphasis on symmetry and boundedness of the distances. Two histogram generation methods based on Gauss mixture vector quantization (GMVQ) are compared using Kullback-Leibler divergence (KLD). The joint histogram method shows a better retrieval performance than the Bayesian retrieval with the label histograms of interleaved data. A variety of distance measures are tested and compared for the joint histogram features produced by GMVQ, including an important set of Ali-Silvey distances, the Bhattacharyya distance, and a few other divergence measures based on Shannon entropy. Experimental results show that the Bhattacharyya distance and the L divergence are better than the histogram intersection (HI), but the KLD is poorer than the HI. In all cases, the symmetric version of a distance performs better than the asymmetric one and usually the bounded version of a distance gives better retrieval performance than the corresponding non-bounded.