Presentation

<
GAIA: Generative Animatable Interactive Avatars with Expression-conditioned Gaussians
SessionThe Shape of You
DescriptionWe present GAIA (Generative Animatable Interactive Avatars) for high-fidelity 3D head avatar generation. GAIA learns dynamic details with expression-conditioned Gaussians, while being animatable consistently with an underlying morphable model. With a novel two-branch architecture, GAIA disentangles identity and expression. GAIA achieves state-of-the-art realism and supports interactive rendering and animation.
Authors
Zhengming Yu
Texas A&M University
Tianye Li
NVIDIA
Jingxiang Sun
Tsinghua University
NVIDIA
Omer Shapira
NVIDIA
Seonwook Park
NVIDIA
Michael Stengel
NVIDIA
Matthew Chan
NVIDIA
Xin Li
Texas A&M University
Wenping Wang
Texas A&M University
Koki Nagano
NVIDIA
Shalini De Mello
NVIDIA
Event Type
Technical Paper
TimeThursday, 14 August 20259:00am - 9:10am PDT
LocationWest Building, Rooms 211-214
ACM Digital Library Journal Papers' PDFs
Conference Papers' PDFs
Session Time & Location
Sunday, 10 August 20256:00pm - 8:45pm PDTWest Building, Ballroom AB
Thursday, 14 August 20259:00am - 10:30am PDTWest Building, Rooms 211-214
Recordings
Livestreamed
Recorded
Next PresentationNext PresentationTeGA: Texture Space Gaussian Avatars for High-Resolution Dynamic Head Modeling
Similar Presentations
Emerging Technologies Demo - GAIA: Generative Animatable Interactive Avatars with Expression-conditioned Gaussians
Text-based Animatable 3D Avatars with Morphable Model Alignment
LAM: Large Avatar Model for One-shot Animatable Gaussian Head