NerfAcc: Efficient Sampling Accelerates NeRFs

TL;DR

NerfAcc introduces a flexible sampling framework that accelerates NeRF training significantly, enabling easier integration of advanced sampling methods and reducing computational costs across various NeRF variants.

Contribution

The paper presents NerfAcc, a Python toolbox that simplifies incorporating improved sampling strategies into NeRFs, leading to substantial training speedups and broader applicability.

Findings

Training time reduced by up to 20x

Applicable across multiple NeRF variants

Supports native PyTorch implementation

Abstract

Optimizing and rendering Neural Radiance Fields is computationally expensive due to the vast number of samples required by volume rendering. Recent works have included alternative sampling approaches to help accelerate their methods, however, they are often not the focus of the work. In this paper, we investigate and compare multiple sampling approaches and demonstrate that improved sampling is generally applicable across NeRF variants under an unified concept of transmittance estimator. To facilitate future experiments, we develop NerfAcc, a Python toolbox that provides flexible APIs for incorporating advanced sampling methods into NeRF related methods. We demonstrate its flexibility by showing that it can reduce the training time of several recent NeRF methods by 1.5x to 20x with minimal modifications to the existing codebase. Additionally, highly customized NeRFs, such as Instant-NGP, can be implemented in native PyTorch using NerfAcc.