.png){kind=icon}
Return
Categorical Codebook Matching for Embodied Character Controllers
.png){kind=icon}
PRE
AI
DOI：10.1145/3658209.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Translating motions from a real user onto a virtual embodied avatar is a key challenge for character animation in the metaverse. In this work, we present a novel generative framework that enables mapping from a set of sparse sensor signals to a full body avatar motion in real-time while faithfully preserving the motion context of the user. In contrast to existing techniques that require training a motion prior and its mapping from control to motion separately, our framework is able to learn the motion manifold as well as how to sample from it at the same time in an end-to-end manner. To achieve that, we introduce a technique called codebook matching which matches the probability distribution between two categorical codebooks for the inputs and outputs for synthesizing the character motions. We demonstrate this technique can successfully handle ambiguity in motion generation and produce high quality character controllers from unstructured motion capture data. Our method is especially useful for interactive applications like virtual reality or video games where high accuracy and responsiveness are needed.
Keywords:
neural networks
human motion
character animation
character control
character interactions
deep learning
Journal
IF:
9.5
Papers:
4.8K
Citations:
3.6W
Organization
Cited Papers
ShareSave
ShareSaveEffects of oestrone and progesterone on collagen and ascorbic acid content in the endometrium and myometrium of the rat.png)
Reproduction
OAAI
ReproductionIF0
OAAIShareSaveShareSaveShareSaveShareSave
View more