SaLF: Sparse Local Fields for Multi-Sensor Rendering in Real-Time

TL;DR

SaLF introduces a unified, efficient volumetric representation supporting multi-sensor simulation with fast training and rendering, enabling scalable and realistic autonomous sensor testing.

Contribution

SaLF is a novel sparse volumetric representation that supports both rasterization and raytracing for multi-sensor simulation, improving efficiency and flexibility.

Findings

SaLF achieves training in under 30 minutes.

SaLF renders at over 50 FPS for cameras and 600 FPS for LiDAR.

SaLF supports non-pinhole cameras and spinning LiDARs.

Abstract

High-fidelity sensor simulation of light-based sensors such as cameras and LiDARs is critical for safe and accurate autonomy testing. Neural radiance field (NeRF)-based methods that reconstruct sensor observations via ray-casting of implicit representations have demonstrated accurate simulation of driving scenes, but are slow to train and render, hampering scalability. 3D Gaussian Splatting (3DGS) has demonstrated faster training and rendering times through rasterization, but is primarily restricted to pinhole camera sensors, preventing usage for realistic multi-sensor autonomy evaluation. Moreover, both NeRF and 3DGS couple the representation with the rendering procedure (implicit networks for ray-based evaluation, particles for rasterization), preventing interoperability, which is key for general usage. In this work, we present Sparse Local Fields (SaLF), a novel volumetric representation that supports rasterization and raytracing for unified multi-sensor simulation. SaLF represents volumes as a sparse set of 3D voxel primitives, where each voxel is a local implicit field. SaLF has fast training ($<$30 min) and rendering capabilities (50+ FPS for camera and 600+ FPS for LiDAR), has adaptive pruning and densification to easily handle large scenes, and can support non-pinhole cameras and spinning LiDARs. We demonstrate that SaLF has similar realism as existing self-driving sensor simulation methods while improving efficiency and enhancing capabilities, enabling more scalable simulation.