Presentation

<
Categorical Codebook Matching for Embodied Character Controllers
SessionCharacter Control
DescriptionThis work presents a generative VQ framework to translate movements of a real user onto a virtual embodied avatar using VR inputs. Our proposed codebook matching technique enables simultaneously learning and sampling the motions in form of Categorical probabilities and produces realistic full-body motion behaviors, accurately and with zero delay.
Authors
Sebastian Starke
Facebook Reality Labs
Paul Starke
Facebook Reality Labs
Taku Komura
University of Hong Kong
Yuting Ye
Facebook Reality Labs
Event Type
Technical Paper
TimeThursday, 1 August 20242:00pm - 2:10pm MDT
LocationMile High 4
ACM Digital Library Journal Papers' PDFs
Conference Papers' PDFs
Session Time & Location
Sunday, 28 July 20246:00pm - 8:45pm MDTBluebird Ballroom
Thursday, 1 August 20242:00pm - 3:30pm MDTMile High 4
Interest Areas
Research & Education
Recordings
Livestreamed
Recorded
Keywords
Animation
Registration Categories
Full Conference
Full Conference Supporter
Virtual Access
Exhibitor Full Conference
Thursday
Next PresentationNext PresentationInteractive Character Control With Auto-regressive Motion Diffusion Models