Real-Time Radiance Fields for Single-Image Portrait View Synthesis

Abstract

We 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 3D-aware novel view synthesis via volume rendering. Our method is fast (24 fps) on consumer hardware, and produces higher quality results than strong GAN-inversion baselines that require test-time optimization. To train our triplane encoder pipeline, we use only synthetic data, showing how to distill the knowledge from a pretrained 3D GAN into a feedforward encoder. Technical contributions include a Vision Transformer-based triplane encoder, a camera data augmentation strategy, and a well-designed loss function for synthetic data training. We benchmark against the state-of-the-art methods, demonstrating significant improvements in robustness and image quality in challenging real-world settings. We showcase our results on portraits of faces (FFHQ) and cats (AFHQ), but our algorithm can also be applied in the future to other categories with a 3D-aware image generator.

Document Details

Document Type
Pub Defense Publication
Publication Date
Jul 26, 2023
Source ID
10.1145/3592460

Entities

People

  • Alex Trevithick
  • Chao Liu
  • Eric R. Chan
  • Koki Nagano
  • Manmohan Chandraker
  • Matthew Chan
  • Michael Stengel
  • Ravi Ramamoorthi
  • Sameh Khamis
  • Zhiding Yu

Organizations

  • Defense Advanced Research Projects Agency
  • National Science Foundation
  • Nvidia
  • Office of Naval Research
  • Stanford University
  • University of California, San Diego

Tags

Fields of Study

  • Computer science

Readers

  • Brain and Cognitive Science; Experimental Psychology; Cognitive Neuroscience
  • Computer Vision.
  • Systems Analysis and Design

Technology Areas

  • AI & ML
  • AI & ML - Neural Networks