Hits:

Date of Publication:2009-01-01

Journal:自动化学报

Volume:35

Issue:7

Page Number:1016-1021

ISSN No.:0254-4156

Abstract:A mean shift quasi-Monte Carlo (MS-QMC) method is proposed for speaker
   tracking. To explore the state space more efficiently, deterministic
   samplers are used instead of random draws according to a quasi-Monte
   Carlo integration rule in the new method. Furthermore, a mean shift
   procedure is applied to move particles toward the modes of the
   posterior, leading to a more effective allocation of particles thereupon
   fewer particles are needed. Simulation results show that compared with
   the traditional particle filter, both speaker tracking accuracy and
   convergent rate of the proposed method are improved.

Note:新增回溯数据