Presentation / Installation


A Hybrid Generator Architecture for Controllable Face Synthesis
SessionMaking Faces With Neural Avatars
DescriptionThis research presents a data-driven generator architecture that combines the photorealistic qualities of GANs with disentangled controls and 3D mesh manipulation. This approach enables precise control and interpretability without the need for labeled training data, ensuring compatibility with traditional animation tools, and promoting greater flexibility and creativity.
Authors
Dann Mensah
Aalto University
Nam Hee Kim
Aalto University
Miika Aittala
NVIDIA
Samuli Laine
NVIDIA
Jaakko Lehtinen
Aalto University
NVIDIA
Event Type
Technical Paper
TimeThursday, 10 August 202311:26am - 11:36am PDT
LocationPetree Hall D
ACM Digital Library Technical Paper PDF
Session Time & Location
Sunday, 6 August 20236pm - 8:30pm PDTWest Hall B
Thursday, 10 August 202310:45am - 12:15pm PDTPetree Hall D
Interest Areas
Research & Education
Recordings
Livestreamed
Recorded
Keywords
Animation/Simulation
Modeling
Video
Registration Categories
FC
FCS
V
VS
EFC
Next PresentationNext Presentation
ClipFace: Text-guided Editing of Textured 3D Morphable Models