NeRFPlayer: A Streamable Dynamic Scene Representation with Decomposed Neural Radiance Fields

TL;DR

NeRFPlayer introduces a fast, streamable neural radiance field framework for dynamic scene reconstruction and rendering from minimal camera input, enabling real-time exploration in VR.

Contribution

It decomposes 4D space into static, deforming, and new areas with separate neural fields and employs a hybrid feature streaming scheme for efficient modeling.

Findings

Achieves reconstruction in 10 seconds per frame

Provides interactive rendering performance

Works with single or multiple camera setups

Abstract

Visually exploring in a real-world 4D spatiotemporal space freely in VR has been a long-term quest. The task is especially appealing when only a few or even single RGB cameras are used for capturing the dynamic scene. To this end, we present an efficient framework capable of fast reconstruction, compact modeling, and streamable rendering. First, we propose to decompose the 4D spatiotemporal space according to temporal characteristics. Points in the 4D space are associated with probabilities of belonging to three categories: static, deforming, and new areas. Each area is represented and regularized by a separate neural field. Second, we propose a hybrid representations based feature streaming scheme for efficiently modeling the neural fields. Our approach, coined NeRFPlayer, is evaluated on dynamic scenes captured by single hand-held cameras and multi-camera arrays, achieving comparable or superior rendering performance in terms of quality and speed comparable to recent state-of-the-art methods, achieving reconstruction in 10 seconds per frame and interactive rendering.