Presentation 2010-01-22
Mean Shift Tracking based on Feature Selection and Multiple Subtrackers
Kohei MARUYAMA, Takeshi SAITOH, Ryosuke KONISHI,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) This paper proposes the Mean Shift tracking combined feature selection and multiple subtrackers. Our method has three advantages. (1) Robust to the change of the background environment by using feature selection method. (2) Robust to the change of the shape and achieve the accurate modeling by multiple subtrackers. (3) Respond to the rapid scale change to change the number of subtrackers adaptive in each frame. The overview of our method is as follows: the feature selection is applied to select effective feature in each frame based on the parent tracker of previous frame. Then multiple subtrackers are generated. Each subtracker is applied the mean shift tracking. Finally, the elimination and fusion processes are applied and estimate new location. The proposed method confirmed high tracking accuracy was obtained compared with the traditional method though the problem at the processing time was remained.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Mean shift tracking / feature selection / subtrackers
Paper # CQ2009-96,PRMU2009-195,SP2009-136,MVE2009-118
Date of Issue

Conference Information
Committee CQ
Conference Date 2010/1/14(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Communication Quality (CQ)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Mean Shift Tracking based on Feature Selection and Multiple Subtrackers
Sub Title (in English)
Keyword(1) Mean shift tracking
Keyword(2) feature selection
Keyword(3) subtrackers
1st Author's Name Kohei MARUYAMA
1st Author's Affiliation Tottori University()
2nd Author's Name Takeshi SAITOH
2nd Author's Affiliation Tottori University
3rd Author's Name Ryosuke KONISHI
3rd Author's Affiliation Tottori University
Date 2010-01-22
Paper # CQ2009-96,PRMU2009-195,SP2009-136,MVE2009-118
Volume (vol) vol.109
Number (no) 373
Page pp.pp.-
#Pages 6
Date of Issue