A Fully Interpretable Statistical Approach for Roadside LiDAR Background Subtraction

TL;DR

This paper introduces a fully interpretable statistical method using a Gaussian distribution grid for background subtraction in roadside LiDAR data, improving accuracy and flexibility for automated driving perception.

Contribution

It proposes a novel Gaussian distribution grid model and filtering algorithm that are adaptable to various LiDAR types and configurations, enhancing background subtraction performance.

Findings

Outperforms state-of-the-art methods in accuracy on RCooper dataset

Supports diverse LiDAR sensors including MEMS and 360-degree systems

Operates efficiently on low-resource hardware

Abstract

We present a fully interpretable and flexible statistical method for background subtraction in roadside LiDAR data, aimed at enhancing infrastructure-based perception in automated driving. Our approach introduces both a Gaussian distribution grid (GDG), which models the spatial statistics of the background using background-only scans, and a filtering algorithm that uses this representation to classify LiDAR points as foreground or background. The method supports diverse LiDAR types, including multiline 360 degree and micro-electro-mechanical systems (MEMS) sensors, and adapts to various configurations. Evaluated on the publicly available RCooper dataset, it outperforms state-of-the-art techniques in accuracy and flexibility, even with minimal background data. Its efficient implementation ensures reliable performance on low-resource hardware, enabling scalable real-world deployment.