Animating Virtual Signers

This paper presents an ongoing PhD research project on visual perception and motion analysis applied to virtual signers (virtual agents used for Sign Language interaction). Virtual signers (or signing avatars) play an important role in the accesibility of information in sign languages. They have been developed notably for their capability to anonymize shape and ap-pearance of the content producer. While motion capture provides human-like, realistic and comprehensible signing animations, it also arises the question of anonymity. Human body movements contain important information about a person's identity, gender or emotional state. In the present work, we want to address the problem of gestural identity in the context of animated agents in French Sign Language. On the one hand, the ability to identify a person from signing motion is assessed through psychophysical experiments, using point-light displays. On the other hand, a computational framework is developed in order to investigate which features are critical for person identification and to control them over the virtual agent.