“Bodyformer: Semantics-guided 3D Body Gesture Synthesis With Transformer” by Pang, Qin, Fan, Habekost, Shiratori, et al. …
Conference:
Type:
Title:
- Bodyformer: Semantics-guided 3D Body Gesture Synthesis With Transformer
Session/Category Title: Character Animation: Knowing What To Do With Your Hands
Presenter(s)/Author(s):
Moderator(s):
Abstract:
This paper presents a new variational transformer framework for synthesizing 3D body gestures driven by speech. The system uses a mode position embedding and intra-modal pre-training to learn motion patterns from limited conversational data. Extensive studies show that the system can generate realistic and diverse gestures similar to the ground-truth.
Additional Images:
![©Kunkun Pang, Dafei Qin, Yingruo Fan, Julian Habekost, Takaaki Shiratori, Junichi Yamagishi, and Taku Komura](https://history.siggraph.org/wp-content/uploads/2024/02/2023-Tech-Papers-Pang_Semantics-guided-3D-Body-Gesture-Synthesis-with-Transformer-02.jpg)