Pedestrian classification from moving platforms using cyclic motion pattern

TitlePedestrian classification from moving platforms using cyclic motion pattern
Publication TypeConference Papers
Year of Publication2005
AuthorsRan Y, Zheng Q, Weiss I, Davis LS, Abd-Almageed W, Zhao L
Conference NameIEEE International Conference on Image Processing, 2005. ICIP 2005
Date Published2005/09//
ISBN Number0-7803-9134-9
Keywordscompact shape representation, cyclic motion pattern, data mining, Detectors, digital phase locked loop, digital phase locked loops, feedback loop module, gait analysis, gait phase information, human body pixel oscillations, HUMANS, image classification, Image motion analysis, image representation, image sequence, Image sequences, Motion detection, Object detection, pedestrian classification, pedestrian detection system, Phase estimation, Phase locked loops, principle gait angle, SHAPE, tracking, Videos

This paper describes an efficient pedestrian detection system for videos acquired from moving platforms. Given a detected and tracked object as a sequence of images within a bounding box, we describe the periodic signature of its motion pattern using a twin-pendulum model. Then a principle gait angle is extracted in every frame providing gait phase information. By estimating the periodicity from the phase data using a digital phase locked loop (dPLL), we quantify the cyclic pattern of the object, which helps us to continuously classify it as a pedestrian. Past approaches have used shape detectors applied to a single image or classifiers based on human body pixel oscillations, but ours is the first to integrate a global cyclic motion model and periodicity analysis. Novel contributions of this paper include: i) development of a compact shape representation of cyclic motion as a signature for a pedestrian, ii) estimation of gait period via a feedback loop module, and iii) implementation of a fast online pedestrian classification system which operates on videos acquired from moving platforms.