Database guided computer animation of human grasping using forward and inverse kinematics

This paper addresses the important issue of automating grasping movement in the animation of virtual actors, and presents a methodology and algorithm to generate realistic looking grasping motion of arbitrary shaped objects. A hybrid approach using both forward and inverse kinematics is proposed. A database of predefined body postures and hand trajectories are generalized to adapt to a specific grasp. The reachable space is divided into small sub volumes, which enables the construction of the database. The paper also add resses some common problems of articulated figure animation. A new approach for body positioning with kinematic constraints on both hands is described . An efficient and accurate manipulation of joint constraints is also prese nted. Finally, we describe an interpolation algorithm which interpolates be tween two postures of an articulated figure by moving the end effector alon g a specific trajectory and maintaining all the joint angles in the feasibl e range. Results are quite satisfactory, and some are shown in the paper. ( C) 1999 Elsevier Science Ltd. All rights reserved.