Video games and virtual reality are becoming commonplace and hand gestures play an important role in the realism of an action or animation. However, production of these animations is still a long and arduous task. The goal of this research is to propose and implement image- and video-based algorithms for the capture and tracking of hand motion (skin, wrinkles, joints, etc.) to allow reconstruction in a video game context as well as retargeting onto another hand. By capturing fine details of the hand's shape and motion, both the ease of creation and the realism of hand animations will be greatly enhanced.

In this context, as part of the national project PlayALL managed by Cap Digital plans, we plan to develop a hand motion capture system. This project encompasses many studios, industrial partners, editors, and research laboratories including LIRIS, the Lyon Research Center for Images and Intelligent Information Systems.