Dynamic Object Detection in Range data using Spatiotemporal Normals

TL;DR

This paper introduces a simple yet effective method for detecting dynamic objects in range data by computing spatiotemporal normals, applicable to LiDAR and depth cameras, matching state-of-the-art performance with reduced complexity.

Contribution

The paper proposes a novel, straightforward approach using spatiotemporal normals for dynamic object detection in point clouds, simplifying existing methods.

Findings

Robust detection of dynamic objects using spatiotemporal normals

Performance comparable to state-of-the-art methods

Applicable to LiDAR and depth camera data

Abstract

On the journey to enable robots to interact with the real world where humans, animals, and unpredictable elements are acting as independent agents; it is crucial for robots to have the capability to detect dynamic objects. In this paper, we argue that the detection of dynamic objects can be solved by computing the spatiotemporal normals of a point cloud. In our experiments, we demonstrate that this simple method can be used robustly for LiDAR and depth cameras with performances similar to the state of the art while offering a significantly simpler method.