How Far Can We Compress Instant-NGP-Based NeRF?

TL;DR

This paper introduces a novel context-based compression framework for NeRF that significantly reduces storage size while maintaining high fidelity, leveraging context models and prior knowledge.

Contribution

It is the first to construct and exploit context models specifically for NeRF compression, achieving substantial size reduction with improved fidelity.

Findings

Achieves 100x and 70x size reduction on benchmark datasets.

Outperforms state-of-the-art NeRF compression methods in storage efficiency.

Maintains high rendering fidelity despite compression.

Abstract

In recent years, Neural Radiance Field (NeRF) has demonstrated remarkable capabilities in representing 3D scenes. To expedite the rendering process, learnable explicit representations have been introduced for combination with implicit NeRF representation, which however results in a large storage space requirement. In this paper, we introduce the Context-based NeRF Compression (CNC) framework, which leverages highly efficient context models to provide a storage-friendly NeRF representation. Specifically, we excavate both level-wise and dimension-wise context dependencies to enable probability prediction for information entropy reduction. Additionally, we exploit hash collision and occupancy grids as strong prior knowledge for better context modeling. To the best of our knowledge, we are the first to construct and exploit context models for NeRF compression. We achieve a size reduction of 100$\times$ and 70$\times$ with improved fidelity against the baseline Instant-NGP on Synthesic-NeRF and Tanks and Temples datasets, respectively. Additionally, we attain 86.7\% and 82.3\% storage size reduction against the SOTA NeRF compression method BiRF. Our code is available here: https://github.com/YihangChen-ee/CNC.