Presentation
Bringing Diversity from Diffusion Models to Semantic-Guided Face Asset Generation
SessionFaces & Avatars
DescriptionThis paper introduces a framework for generating high-quality, semantically controllable facial assets. By leveraging a diffusion model's diversity, we develop a face asset synthesis pipeline to produce paired albedo and geometry data with semantic labels. These pairs are then used to train a disentangled GAN, enabling precise latent editing of the generated assets while preserving identity.
Event Type
Technical Paper
TimeMonday, 20 July 20262:30pm - 2:40pm PDT
LocationRoom 403 A
Session Time & Location
Sunday, 19 July 20266:00pm - 8:45pm PDTHall K
Monday, 20 July 20262:00pm - 3:30pm PDTRoom 403 A
Geometry
Modeling
Full Conference Supporter
Full Conference
