Skip to main content
  1. Papers/

A-NeRF

A-NeRF: Articulated Neural Radiance Fields for Learning Human Shape, Appearance, and Pose
#

Shih-Yang Su, Frank Yu, Michael Zollhöfer, Helge Rhodin

NeRF
Skeleton
NeurIPS 2021
LemonATsu/A-NeRF

Pytorch implementation for A-NeRF: Articulated Neural Radiance Fields for Learning Human Shape, Appearance, and Pose

Python
199
26

Abstract
#

While deep learning reshaped the classical motion capture pipeline with feedforward networks, generative models are required to recover fine alignment via iterative refinement. Unfortunately, the existing models are usually hand-crafted or learned in controlled conditions, only applicable to limited domains. We propose a method to learn a generative neural body model from unlabelled monocular videos by extending Neural Radiance Fields (NeRFs). We equip them with a skeleton to apply to time-varying and articulated motion. A key insight is that implicit models require the inverse of the forward kinematics used in explicit surface models. Our reparameterization defines spatial latent variables relative to the pose of body parts and thereby overcomes ill-posed inverse operations with an overparameterization. This enables learning volumetric body shape and appearance from scratch while jointly refining the articulated pose; all without ground truth labels for appearance, pose, or 3D shape on the input videos. When used for novel-view-synthesis and motion capture, our neural model improves accuracy on diverse datasets.
Paper

Approach
#

A-NeRF overview
A-NeRF overview.

Results
#

Data
#

Comparisons
#

Performance
#

Papers Published @ 2021 - This article is part of a series.
Part 1: This Article