Streaming Radiance Fields for 3D Video Synthesis

TL;DR

This paper introduces a streaming radiance field method for real-time 3D video synthesis that significantly accelerates training and reduces storage, enabling fast on-the-fly scene reconstruction with high quality.

Contribution

It proposes an explicit-grid based incremental learning framework with model difference compression and efficient optimization strategies for dynamic scene reconstruction.

Findings

Achieves 15 seconds per-frame training speed.

Provides 1000x speedup over implicit methods.

Maintains competitive rendering quality.

Abstract

We present an explicit-grid based method for efficiently reconstructing streaming radiance fields for novel view synthesis of real world dynamic scenes. Instead of training a single model that combines all the frames, we formulate the dynamic modeling problem with an incremental learning paradigm in which per-frame model difference is trained to complement the adaption of a base model on the current frame. By exploiting the simple yet effective tuning strategy with narrow bands, the proposed method realizes a feasible framework for handling video sequences on-the-fly with high training efficiency. The storage overhead induced by using explicit grid representations can be significantly reduced through the use of model difference based compression. We also introduce an efficient strategy to further accelerate model optimization for each frame. Experiments on challenging video sequences demonstrate that our approach is capable of achieving a training speed of 15 seconds per-frame with competitive rendering quality, which attains $1000 \times$ speedup over the state-of-the-art implicit methods. Code is available at https://github.com/AlgoHunt/StreamRF.