Presentation 2000/1/21
Tracking pedestrians in the crowd
Kyousuke Uchida, Jun Miura, Yoshiaki Shirai,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) For the city designing research, we need to know how pedestrians make way or which shop they often go in. This paper describes the algorithm of tracking all pedestrians by a fixed camera, which slantwise looks down a pedestrian precinct. At every frame pedestrians are extracted by the background image subtraction method and tracked forward and backward. If a locus branches into two locusci, both of them are tracked until the ambiguity is resolved. We can get the locusci of pedestrians robustly even if they overlap in the crowd, by detecting each of them at a certain frame and by tracking each of them forward and backward.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) tracking / locus / detection / sequence
Paper # PRMU99-209
Date of Issue

Conference Information
Committee PRMU
Conference Date 2000/1/21(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 Pattern Recognition and Media Understanding (PRMU)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Tracking pedestrians in the crowd
Sub Title (in English)
Keyword(1) tracking
Keyword(2) locus
Keyword(3) detection
Keyword(4) sequence
1st Author's Name Kyousuke Uchida
1st Author's Affiliation Dept. of Computer-Controlled Mechanical Systems, Osaka University()
2nd Author's Name Jun Miura
2nd Author's Affiliation Dept. of Computer-Controlled Mechanical Systems, Osaka University
3rd Author's Name Yoshiaki Shirai
3rd Author's Affiliation Dept. of Computer-Controlled Mechanical Systems, Osaka University
Date 2000/1/21
Paper # PRMU99-209
Volume (vol) vol.99
Number (no) 575
Page pp.pp.-
#Pages 6
Date of Issue