HomeResearchPublicationsContact

 

NATURAL INTERACTION PROTOTYPES BY MEANS OF ENACTIVE
INTERFACES BASED ON VISUAL INPUTS

Research – Pose estimation

postura3In this research we explore the recovery of the full body posture of a standing subject in front of a stereo camera system. The 3D position of the hands, the head and the center of the trunk egment are extracted in real-time and provided to the body posture recovery algorithmic layer. However, locating all the user body joints to recover the posture is not possible with only computer vision algorithms. This is mainly due to the fact that most of the joints are occluded by clothes. The idea is to recover a believable posture from the position of certain joints, the end-effectors. This is defined as Inverse Kinematics (IK), also called ‘goal-directed motion’ in computer animation, where the animator only defines the position of the end-effectors to achieve a desired motion. IK solves for the position and orientation of all joints in the link hierarchy that lead to such end-effectors.
In the general case there is no analytic solution for the IK problem. However, IK can be solved via non-linear programming techniques. In our case, the multiple Priority IK (also called Prioritized IK or PIK) is exploited to reconstruct an anatomically correct posture of the user considering the 3D locations of selected end-effectors which are measured with the vision system and are used to constrain the posture.

 

postura0

References

Boulic, R.; Varona, J.; Unzueta, L.; Peinado, M.; Suescun, A.; Perales, F.J.,
“Evaluation of on-line analytic and numeric inverse kinematics approaches driven
by partial vision input”. Virtual Reality 10(1): 48-61, 2006.

Application: Collision detection in Virtual Environments.