Generative Neural Articulated Radiance Fields

TL;DR

This paper introduces a novel 3D GAN framework capable of generating and editing high-quality radiance fields of human bodies and faces in various poses, advancing 3D-aware generative modeling.

Contribution

It presents the first 3D GAN for human body radiance fields with explicit deformation-based pose editing, improving quality over previous methods.

Findings

First high-quality 3D human body radiance field generation.

Deformation-aware training enhances pose and expression editing.

Framework enables direct editing of generated 3D human models.

Abstract

Unsupervised learning of 3D-aware generative adversarial networks (GANs) using only collections of single-view 2D photographs has very recently made much progress. These 3D GANs, however, have not been demonstrated for human bodies and the generated radiance fields of existing frameworks are not directly editable, limiting their applicability in downstream tasks. We propose a solution to these challenges by developing a 3D GAN framework that learns to generate radiance fields of human bodies or faces in a canonical pose and warp them using an explicit deformation field into a desired body pose or facial expression. Using our framework, we demonstrate the first high-quality radiance field generation results for human bodies. Moreover, we show that our deformation-aware training procedure significantly improves the quality of generated bodies or faces when editing their poses or facial expressions compared to a 3D GAN that is not trained with explicit deformations.