Multiple Hypothesis Tracking for Automatic Optical Motion Capture
ECCV '02 Proceedings of the 7th European Conference on Computer Vision-Part I
Monocuolar Perception of Biological Motion - Clutter and Partial Occlusion
ECCV '00 Proceedings of the 6th European Conference on Computer Vision-Part II
Automatic acquisition and initialization of articulated models
Machine Vision and Applications - Special issue: Human modeling, analysis, and synthesis
Scene-consistent detection of feature points in video sequences
Computer Vision and Image Understanding
The emergence of visual categories: a computational perspective
SCIA'03 Proceedings of the 13th Scandinavian conference on Image analysis
| Hi-index | 0.00 |
Computer perception of biological motion is key to developing convenient and powerful human-computer inter-faces. Successful body tracking algorithms have been developed; however, initialization is done by hand.We propose a method for detecting a moving human body and for labeling its parts automatically. It is based on maximizing the joint probability density function (PDF) of the position and velocity of the body parts. The PDF is estimated from training data. Dynamic programming is used for calculating efficiently the best global labeling on an approximation of the PDF. The computational cost is on the order of N 4 where N is the number of features detected.We explore the performance of our method with experiments carried on a variety of periodic and non-periodic body motions viewed monocularly for a total of approximately 30,000 frames. Point-markers were strapped to the joints of the subject for facilitating image analysis. We find an average of 2.3% labeling error; the experiments also suggest a high degree of viewpoint-invariance.