A Game Theory-Based Multi-Feature Fusion for Object Tracking
JIN Zefenfen
HOU Zhiqiang
YU Wangsheng
WANG Xin
JIA Yuhu
Abstract：Aimed on the problem that the robustness of single feature tracking algorithm is poor in visual target tracking,this paper proposes an object tracking algorithm based on multi-feature fusion using game theory.Under condition of the mean shift tracking framework,the paper takes color feature and motion feature as two players.Through looking for the Nash equilibrium of their game,the paper makes the contribution of different features in the tracking result the best balance,furthermore a higher advantage of multi-feature fusion.The experimental results show that this algorithm has the stronger robustness of tracking under object strenuous motion,occlusion and background motion interference.By means of multifeature fusion based on game theory,the paper presents a new algorithm on the basis of the traditional mean shift algorithm,and the algorithm is good in performance on tracking.
Keywords：vision trackingmulti-feature fusiongame theoryMean Shift
Publication Date：2017-01-01
Online Publishing Date：2025-08-15(First online date of this platform, not the publication date of the document)
Pages：7( 50-56 )