Title :
On the sustained tracking of human motion
Author :
Sheikh, Yaser Ajmal ; Datta, Ankur ; Kanade, Takeo
Author_Institution :
Robot. Inst., Carnegie Mellon Univ., Pittsburgh, PA
Abstract :
In this paper, we propose an algorithm for sustained tracking of humans, where we combine frame-to-frame articulated motion estimation with a per-frame body detection algorithm. The proposed approach can automatically recover from tracking error and drift. The frame-to-frame motion estimation algorithm replaces traditional dynamic models within a filtering framework. Stable and accurate per-frame motion is estimated via an image-gradient based algorithm that solves a linear constrained least squares system. The per-frame detector learns appearance of different body parts and dasiasketchespsila expected gradient maps to detect discriminant pose configurations in images. The resulting online algorithm is computationally efficient and has been widely tested on a large dataset of sequences of drivers in vehicles. It shows stability and sustained accuracy over thousands of frames.
Keywords :
filtering theory; gradient methods; least squares approximations; motion estimation; target tracking; filtering framework; frame-to-frame articulated motion estimation; image-gradient based algorithm; linear constrained least squares system; per-frame body detection algorithm; sustained human motion tracking; Biological system modeling; Detection algorithms; Detectors; Filtering algorithms; Heuristic algorithms; Humans; Least squares approximation; Motion estimation; Tracking; Vehicle dynamics;
Conference_Titel :
Automatic Face & Gesture Recognition, 2008. FG '08. 8th IEEE International Conference on
Conference_Location :
Amsterdam
Print_ISBN :
978-1-4244-2153-4
Electronic_ISBN :
978-1-4244-2154-1
DOI :
10.1109/AFGR.2008.4813456