Physically Based Neural LiDAR Resimulation

TL;DR

This paper introduces a physically based neural LiDAR resimulation method that explicitly models sensor effects for more accurate and high-resolution LiDAR data generation, advancing the realism and utility of LiDAR simulations.

Contribution

It presents a novel neural LiDAR resimulation approach that explicitly incorporates sensor effects, improving accuracy and enabling high-resolution resimulation from camera perspectives.

Findings

Outperforms existing LiDAR simulation methods in accuracy.

Effectively models sensor effects like rolling shutter and laser power.

Enables high-resolution LiDAR resimulation from camera views.

Abstract

Methods for Novel View Synthesis (NVS) have recently found traction in the field of LiDAR simulation and large-scale 3D scene reconstruction. While solutions for faster rendering or handling dynamic scenes have been proposed, LiDAR specific effects remain insufficiently addressed. By explicitly modeling sensor characteristics such as rolling shutter, laser power variations, and intensity falloff, our method achieves more accurate LiDAR simulation compared to existing techniques. We demonstrate the effectiveness of our approach through quantitative and qualitative comparisons with state-of-the-art methods, as well as ablation studies that highlight the importance of each sensor model component. Beyond that, we show that our approach exhibits advanced resimulation capabilities, such as generating high resolution LiDAR scans in the camera perspective. Our code and the resulting dataset are available at https://github.com/richardmarcus/PBNLiDAR.