Tracking and recognizing two-person interactions in outdoor image sequences

Author

Sato, Koichi ; Aggarwal, J.K.

Author_Institution

Dept. of Electr. & Comput. Eng., Texas Univ., Austin, TX, USA

fYear

2001

fDate

2001

Firstpage

87

Lastpage

94

Abstract

This paper presents a methodology for tracking persons and identifying two-person interactions in outdoor image sequences. By locating and tracking two persons over consecutive frames of monocular grayscale image sequences, we classify their interactions into several classes. Some of the interaction classes are: one person leaves another stationary person; two persons meet coming from different directions; and one stationary person starts following another walking person. We use side-view image sequences obtained by a fixed camera. In these image sequences the subjects are frequently occluded and move perpendicular to the direction of the camera. We present results on tracking of persons and their interactions

Keywords

feature extraction; image classification; image motion analysis; image recognition; image segmentation; image sequences; tracking; camera; feature extraction; human segmentation; image frames; interactions classification; monocular grayscale image sequences; occlusion; outdoor image sequences; side-view image sequences; stationary person; surveillance applications; two-person interactions recognition; two-person interactions tracking; walking person; Cameras; Head; Humans; Image recognition; Image segmentation; Image sequences; Legged locomotion; Principal component analysis; Surveillance; Tracking;

fLanguage

English

Publisher

ieee

Conference_Titel

Multi-Object Tracking, 2001. Proceedings. 2001 IEEE Workshop on

Conference_Location

Vancouver, BC

Print_ISBN

0-7695-1171-6

Type

conf

DOI

10.1109/MOT.2001.937986

Filename

937986