Deep Person Detection in 2D Range Data

TL;DR

This paper introduces an improved deep learning method for detecting humans using 2D laser range data, demonstrating superior performance on a large, real-world dataset compared to existing approaches.

Contribution

The paper extends the DROW detector to human detection, incorporates a temporal window for better accuracy, and provides the largest annotated dataset for 2D range data person detection.

Findings

DROW outperforms current state-of-the-art methods

Adding a temporal window improves detection accuracy

The dataset is the largest of its kind with diverse real-world data

Abstract

Detecting humans is a key skill for mobile robots and intelligent vehicles in a large variety of applications. While the problem is well studied for certain sensory modalities such as image data, few works exist that address this detection task using 2D range data. However, a widespread sensory setup for many mobile robots in service and domestic applications contains a horizontally mounted 2D laser scanner. Detecting people from 2D range data is challenging due to the speed and dynamics of human leg motion and the high levels of occlusion and self-occlusion particularly in crowds of people. While previous approaches mostly relied on handcrafted features, we recently developed the deep learning based wheelchair and walker detector DROW. In this paper, we show the generalization to people, including small modifications that significantly boost DROW's performance. Additionally, by providing a small, fully online temporal window in our network, we further boost our score. We extend the DROW dataset with person annotations, making this the largest dataset of person annotations in 2D range data, recorded during several days in a real-world environment with high diversity. Extensive experiments with three current baseline methods indicate it is a challenging dataset, on which our improved DROW detector beats the current state-of-the-art.