Evolving optimal feature set by interactive reinforcement learning for image retrieval

Authors:
Jianbo Su;Fang Liu;Zhiwei Luo
Affiliations:
Department of Automation, Shanghai Jiao Tong University, Shanghai, China and Bio-Mimetic Control Research Center, RIKEN, Moriyama-ku, Nagoya, Japan;Department of Automation, Shanghai Jiao Tong University, Shanghai, China;Bio-Mimetic Control Research Center, RIKEN, Moriyama-ku, Nagoya, Japan
Venue:
ISNN'05 Proceedings of the Second international conference on Advances in neural networks - Volume Part II
Year:
2005

Citing 5
Cited 0

Support vector machine active learning for image retrieval

MULTIMEDIA '01 Proceedings of the ninth ACM international conference on Multimedia
Introduction to Reinforcement Learning

Introduction to Reinforcement Learning
Image Information Systems: Where Do We Go From Here?

IEEE Transactions on Knowledge and Data Engineering
The Bayesian image retrieval system, PicHunter: theory, implementation, and psychophysical experiments

IEEE Transactions on Image Processing
Relevance feedback: a power tool for interactive content-based image retrieval

IEEE Transactions on Circuits and Systems for Video Technology

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper proposes a new image retrieval strategy based on the optimal feature subset that is iteratively learned from the query image. The optimal feature set that can well describe the essential properties of the query image with respect to a retrieved image database is obtained from reinforcement learning procedure with the help of humancomputer interaction. Through human-computer interaction, user can provide similarity evaluation between the query and retrieved images, which actually gives the relevance feedback for a contend-based image retrieval method, and further serves as environmental rewards to feature set evolution actions in reinforcement learning procedure. Experiment results show the effectiveness of the proposed method.