Presentation / Installation


Efficient Video Portrait Reenactment via Grid-based Codebook
SessionMaking Faces With Neural Avatars
DescriptionWe propose the Video Portrait via Grid-based Codebook (VPGC) framework to achieve efficient and high-fidelity portrait modeling. Our key insight is to query driving signals from a position-aware textural codebook with an explicit grid structure. The learned grid-based codebook stores delicate textural information and enables more accurate driving performance.
Authors
Kaisiyuan Wang
The University of Sydney
Hang Zhou
Baidu Inc.
Qianyi Wu
Monash University
Jiaxiang Tang
Peking University
Zhiliang Xu
Baidu Inc.
Borong Liang
Baidu Inc.
Tianshu Hu
Baidu Inc.
Errui Ding
Baidu Inc.
Jingtuo Liu
Baidu Inc.
Ziwei Liu
Nanyang Technological University
Jingdong Wang
Baidu Inc.
Event Type
Technical Paper
TimeThursday, 10 August 202310:56am - 11:06am 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
StyleAvatar: Real-time Photo-realistic Neural Portrait Avatar From a Single Video