Enrico Di Bernardo
University of Padova,
Padova, Italy
Observing the human body in motion is key to a large number of activities and applications such as security, character animation, virtual reality, human-machine interfaces, biomechanics studies, signaling in noisy environments, camera control, traffic and customer monitoring. All of the current techniques for tracking the human body require either employing dedicated human operators or using ad-hoc sensors. This results in a number of limitations: practicality (the user needs to wear markers or other ad-hoc equipment which may be impractical, uncomfortable, constrain the user to a limited work space, be difficult to transport),cost (computational and sensory hardware and human operator time), timeliness (the data may not be available in real-time, but only after a lag required to process a batch of images, allow communication between human operators). If tracking the human body could be made automatic and non-invasive, and therefore cheaper, more practical and faster, not only the applications listed above could be better performed, but also a number of new applications would be feasible. In this talk we will present a model-based technique for the recursive estimation of the human body pose in 3D from a monocular sequence of images. We have developed a real-time system which is able to reconstruct the human arm pose in 3D with an 8% error in the direction of the field of view.