Mapless Online Detection of Dynamic Objects in 3D Lidar

TL;DR

This paper introduces a setting-independent, model-free online method for detecting dynamic objects in 3D lidar data, explicitly compensating for motion distortion to improve accuracy in autonomous driving scenarios.

Contribution

It presents a novel motion-compensated detection algorithm and establishes a new benchmark for evaluating dynamic object detection in motion-distorted lidar data.

Findings

Effective motion compensation improves detection accuracy.

Benchmark results demonstrate superiority over existing methods.

Qualitative analysis confirms real-world applicability.

Abstract

This paper presents a model-free, setting-independent method for online detection of dynamic objects in 3D lidar data. We explicitly compensate for the moving-while-scanning operation (motion distortion) of present-day 3D spinning lidar sensors. Our detection method uses a motion-compensated freespace querying algorithm and classifies between dynamic (currently moving) and static (currently stationary) labels at the point level. For a quantitative analysis, we establish a benchmark with motion-distorted lidar data using CARLA, an open-source simulator for autonomous driving research. We also provide a qualitative analysis with real data using a Velodyne HDL-64E in driving scenarios. Compared to existing 3D lidar methods that are model-free, our method is unique because of its setting independence and compensation for pointcloud motion distortion.