Compressing Explicit Voxel Grid Representations: fast NeRFs become also small

Abstract

NeRFs have revolutionized the world of per-scene radiance field reconstruction because of their intrinsic compactness. One of the main limitations of NeRFs is their slow rendering speed, both at training and inference time. Recent research focuses on the optimization of an explicit voxel grid (EVG) that represents the scene, which can be paired with neural networks to learn radiance fields. This approach significantly enhances the speed both at train and inference time, but at the cost of large memory occupation. In this work we propose Re:NeRF, an approach that specifically targets EVG-NeRFs compressibility, aiming to reduce memory storage of NeRF models while maintaining comparable performance. We benchmark our approach with three different EVG-NeRF architectures on four popular benchmarks, showing Re:NeRF's broad usability and effectiveness.

References in corpus (3)

• Instant Neural Graphics Primitives with a Multiresolution Hash Encoding
• LOss-Based SensiTivity rEgulaRization: towards deep sparse neural networks
• SeReNe: Sensitivity based Regularization of Neurons for Structured Sparsity in Neural Networks

Cited by in corpus (5)

• Compact 3D Scene Representation via Self-Organizing Gaussian Grids
• HPC: Hierarchical Progressive Coding Framework for Volumetric Video
• SlimmeRF: Slimmable Radiance Fields
• Optimized 3D Gaussian Splatting using Coarse-to-Fine Image Frequency Modulation
• Learn how to Prune Pixels for Multi-view Neural Image-based Synthesis