An Experimental Study of SOTA LiDAR Segmentation Models

TL;DR

This paper provides a comprehensive experimental comparison of state-of-the-art LiDAR point cloud segmentation models, evaluating their performance, resource usage, and suitability for real-world applications.

Contribution

It offers the first thorough comparison among point-, voxel-, and range image-based models considering practical metrics and application scenarios.

Findings

Voxel-based models achieve higher IoU scores.

Range image-based models have lower inference latency.

Point-based models require less GPU memory.

Abstract

Point cloud segmentation (PCS) is to classify each point in point clouds. The task enables robots to parse their 3D surroundings and run autonomously. According to different point cloud representations, existing PCS models can be roughly divided into point-, voxel-, and range image-based models. However, no work has been found to report comprehensive comparisons among the state-of-the-art point-, voxel-, and range image-based models from an application perspective, bringing difficulty in utilizing these models for real-world scenarios. In this paper, we provide thorough comparisons among the models by considering the LiDAR data motion compensation and the metrics of model parameters, max GPU memory allocated during testing, inference latency, frames per second, intersection-over-union (IoU) and mean IoU (mIoU) scores. The experimental results benefit engineers when choosing a reasonable PCS model for an application and inspire researchers in the PCS field to design more practical models for a real-world scenario.