Presentation
VFAvatar: Feed-Forward 3D Avatar Reconstruction from Casual Image Collections
Session4D Humans
DescriptionWe introduce Visual-Fusion-Avatar (VFAvatar), a novel end-to-end feed-forward framework that reconstructs 3D avatars by fusing visual cues from casual image collections in just a few seconds.
VFAvatar integrates a pose-free reconstruction foundation model with a pretrained human generation prior in a mutually reinforcing manner and achieves SOTA performance.
VFAvatar integrates a pose-free reconstruction foundation model with a pretrained human generation prior in a mutually reinforcing manner and achieves SOTA performance.
Event Type
Technical Paper
TimeWednesday, 22 July 20262:30pm - 2:40pm PDT
LocationRoom 403 A
Digital Library
PDF
Session Time & Location
Sunday, 19 July 20266:00pm - 8:45pm PDTHall K
Wednesday, 22 July 20262:00pm - 3:30pm PDTRoom 403 A
Artificial Intelligence/Machine Learning
Geometry
Modeling
Full Conference Supporter
Full Conference
Similar Presentations
