Title: Sign Language Avatars: Animation and Comprehensibility
Speaker: Alexis Heloir (DFKI)
Time: 2:15pm-3:30pm, Friday, Oct 14
Place: Room 4102, CUNY Graduate Center, 365 Fifth Ave (34str&35str).
		

Abstract:

Many deaf people have significant reading problems. Written
content, e.g. on internet pages, is therefore not fully accessible
for them. Embodied agents have the potential to communicate in the
native language of this cultural group: sign language.

However, state-of-the-art systems have limited comprehensibility
and standard evaluation methods are missing. In this talk, we
present methods and discuss challenges for creating and evaluating
a signing avatar.

We extended an existing general purpose character animation system
with prerequisite functionality, created a gloss-based animation
tool and developed a cyclic content creation workflow with the
help of deaf sign language experts.

For evaluation, we introduce delta testing, a novel way of
assessing comprehensibility by comparing avatars with human
signers. 

To conclude, we present the latest advancements in our animation
system that have the potential to improve the quality and
understandability of future sign language avatars studies.


Speaker's Bio:

Alexis Heloir obtained a masters degree in computer science in 2004 
from the University of Lille I (France). He obtained his PhD degree
from the University of Bretagne Sud in 2008; his thesis examined the 
analysis and synthesis of realistic human motion in regard to French
Sign Language. He is currently post-doctoral researcher at the German
Research Center for Artificial Intelligence (DFKI) in the EMBOTS
research group.  His current research interests include the control 
and the animation of embodied conversational agents as well as the 
generation of intelligible sign language utterances using avatars.