Presentation / Installation


Real-time Radiance Fields for Single-image Portrait View Synthesis
SessionReal-Time Rendering: Gotta Go Fast!
DescriptionWe present a one-shot method to infer and render a photorealistic 3D representation from a single unposed image (e.g., face portrait) in real time. Given a single RGB input, our image encoder directly predicts a canonical triplane representation of a neural radiance field for novel view synthesis via volume rendering.
Authors
Alex Trevithick
UCSD
NVIDIA Research
Matthew Chan
NVIDIA
Michael Stengel
NVIDIA
Eric Chan
Stanford University
Chao Liu
NVIDIA
Zhiding Yu
NVIDIA
Sameh Khamis
NVIDIA
Manmohan Chandraker
UCSD
Ravi Ramamoorthi
UCSD
Koki Nagano
NVIDIA
Event Type
Technical Paper
TimeThursday, 10 August 20239:22am - 9:33am PDT
LocationRoom 515 A
ACM Digital Library Technical Papers pdfs
Session Time & Location
Sunday, 6 August 20236pm - 8:30pm PDTWest Hall B
Thursday, 10 August 20239am - 10:30am PDTRoom 515 A
Interest Areas
Research & Education
Recordings
Livestreamed
Recorded
Keywords
Image Processing
Real-Time
Rendering
Video
Registration Categories
FC
FCS
V
VS
EFC
Next PresentationNext Presentation
Kernel-based Frame Interpolation for Spatio-temporally Adaptive Rendering