A system for human motion matching between synthetic and real images based on a biomechanic graphical model

Abstract

A system for analysis and synthesis of human motion is presented. The system consists of an analysis part and a synthesis part. The analysis part can be used automatically or interactively. The automatic analysis part includes the pre-processing, modeling, matching and interpretation phases. The interactive analysis part includes the same phases but with the possibility of user supervision. We present a global overview of the whole system. The user can define a biomechanic graphical model to represent the human body in a 3D space, and use tools to perform an automatic or interactive supervised matching between the animated model and the real images to recover the motion. The synthesis part uses the results of matching process to show the motion of the human body that is shown in the real images, from any viewpoint. We use specific criteria to match walking persons from different views. Some results and images are presented. >