Natural synthesis of productive forms from structured descriptions of sign language

Abstract

Natural animation of Sign Language directly from linguistic descriptions continues to be a challenge especially in cases where the forms involved are more productive, such as geometric depictions. Prior work laid the foundation for natural Sign Language synthesis with the Paula animation system directly from AZee linguistic descriptions. This paper considers more elaborate discourse, composed of several clauses linked together by the overall meaning and involving largely productive signing. We make the case that one of the keys to natural animation of such discourse lies also in the segments between the typically annotated signs, in other words on the segments traditionally termed "transitions". By studying an example discourse video and the corresponding motion capture, we progressively build an efficient linguistic description of it and specify how to animate it naturally. Sign Language Avatar Transitions Discourse Motion Controls

A Text-To-SL Synthesis System Using 3D Avatar Technology

“…In both cases, sign inflection is only possible at the lexical level. More recently, the Paula system developed at De-Paul University has proven capable of generating animations from the AZee representation, combining techniques based on key postures and procedural algorithms that improve the fluidity of movements and facilitate multimodal synthesis [19,20].…”

Sign Language Avatars: A Question of Representation

“…In contrast, the Zebedee model supports morphological processes through geometric constraints and parametrizable scripts [72]. A subsequent method, AZee, facilitates a more finely detailed and realistic timing of articulator changes, couched in linguistic terms [73].…”