Portrait Neural Radiance Fields from a Single Image
This work provides a method for generating 3D models from single 2D images, which is a significant step towards practical NeRF applications for casual users and moving subjects.
This paper addresses the challenge of creating Neural Radiance Fields (NeRF) from a single portrait image, overcoming NeRF's usual requirement for multiple images. The authors pretrain a multilayer perceptron (MLP) using a meta-learning framework on a light stage portrait dataset and train it in a canonical coordinate space to improve generalization to new faces, achieving favorable results against state-of-the-art methods.
We present a method for estimating Neural Radiance Fields (NeRF) from a single headshot portrait. While NeRF has demonstrated high-quality view synthesis, it requires multiple images of static scenes and thus impractical for casual captures and moving subjects. In this work, we propose to pretrain the weights of a multilayer perceptron (MLP), which implicitly models the volumetric density and colors, with a meta-learning framework using a light stage portrait dataset. To improve the generalization to unseen faces, we train the MLP in the canonical coordinate space approximated by 3D face morphable models. We quantitatively evaluate the method using controlled captures and demonstrate the generalization to real portrait images, showing favorable results against state-of-the-arts.