SCARF: Scalable Continual Learning Framework for Memory-efficiency Multiple Neural Radiance Fields

Wang, Yuze; Wang, Junyi; Wang, Chen; Duan, Wantong; Bao, Yongtang; Qi, Yue

SCARF: Scalable Continual Learning Framework for Memory-efficiency Multiple Neural Radiance Fields

Files

cgf15255.pdf (16.7 MB)

paper1001_mm.pdf (1.74 MB)

Date

2024

Authors

Wang, Yuze
Wang, Junyi
Wang, Chen
Duan, Wantong
Bao, Yongtang
Qi, Yue

Publisher

The Eurographics Association and John Wiley & Sons Ltd.

Abstract

This paper introduces a novel continual learning framework for synthesising novel views of multiple scenes, learning multiple 3D scenes incrementally, and updating the network parameters only with the training data of the upcoming new scene. We build on Neural Radiance Fields (NeRF), which uses multi-layer perceptron to model the density and radiance field of a scene as the implicit function. While NeRF and its extensions have shown a powerful capability of rendering photo-realistic novel views in a single 3D scene, managing these growing 3D NeRF assets efficiently is a new scientific problem. Very few works focus on the efficient representation or continuous learning capability of multiple scenes, which is crucial for the practical applications of NeRF. To achieve these goals, our key idea is to represent multiple scenes as the linear combination of a cross-scene weight matrix and a set of scene-specific weight matrices generated from a global parameter generator. Furthermore, we propose an uncertain surface knowledge distillation strategy to transfer the radiance field knowledge of previous scenes to the new model. Representing multiple 3D scenes with such weight matrices significantly reduces memory requirements. At the same time, the uncertain surface distillation strategy greatly overcomes the catastrophic forgetting problem and maintains the photo-realistic rendering quality of previous scenes. Experiments show that the proposed approach achieves state-of-the-art rendering quality of continual learning NeRF on NeRF-Synthetic, LLFF, and TanksAndTemples datasets while preserving extra low storage cost.

CCS Concepts: Computing methodologies → Rendering; Machine learning; Computer vision

        @article{10.1111:cgf.15255
,
journal = {Computer Graphics Forum},
title = {{SCARF: Scalable Continual Learning Framework for Memory-efficiency Multiple Neural Radiance Fields
}},
author = {Wang, Yuze and 
Wang, Junyi and 
Wang, Chen and 
Duan, Wantong and 
Bao, Yongtang and 
Qi, Yue
},
year = {2024
},
publisher = {The Eurographics Association and John Wiley & Sons Ltd.
},
ISSN = {1467-8659
},
DOI = {10.1111/cgf.15255
}
}