ReLU Fields: The Little Non-linearity That Could
This work addresses the problem of slow training and inference in high-fidelity neural scene representations for computer graphics and vision researchers, offering an incremental improvement over existing methods.
The paper tackles the trade-off between quality and speed in neural scene representations by proposing a simple modification to grid-based methods: applying a fixed ReLU non-linearity to interpolated grid values, which achieves competitive state-of-the-art results in radiance and occupancy fields.
In many recent works, multi-layer perceptions (MLPs) have been shown to be suitable for modeling complex spatially-varying functions including images and 3D scenes. Although the MLPs are able to represent complex scenes with unprecedented quality and memory footprint, this expressive power of the MLPs, however, comes at the cost of long training and inference times. On the other hand, bilinear/trilinear interpolation on regular grid based representations can give fast training and inference times, but cannot match the quality of MLPs without requiring significant additional memory. Hence, in this work, we investigate what is the smallest change to grid-based representations that allows for retaining the high fidelity result of MLPs while enabling fast reconstruction and rendering times. We introduce a surprisingly simple change that achieves this task -- simply allowing a fixed non-linearity (ReLU) on interpolated grid values. When combined with coarse to-fine optimization, we show that such an approach becomes competitive with the state-of-the-art. We report results on radiance fields, and occupancy fields, and compare against multiple existing alternatives. Code and data for the paper are available at https://geometry.cs.ucl.ac.uk/projects/2022/relu_fields.