Pedestrian Detection in 3D Point Clouds using Deep Neural Networks

TL;DR

This paper introduces a deep learning approach using PointNet++ to detect pedestrians solely from 3D LIDAR point clouds, achieving high accuracy without relying on RGB data.

Contribution

It presents a novel method leveraging geometric information from LIDARs with a semi-automatic labeling system for effective pedestrian detection.

Findings

Precision and recall around 98%

Effective use of geometric data alone

Semi-automatic label transfer from RGB images

Abstract

Detecting pedestrians is a crucial task in autonomous driving systems to ensure the safety of drivers and pedestrians. The technologies involved in these algorithms must be precise and reliable, regardless of environment conditions. Relying solely on RGB cameras may not be enough to recognize road environments in situations where cameras cannot capture scenes properly. Some approaches aim to compensate for these limitations by combining RGB cameras with TOF sensors, such as LIDARs. However, there are few works that address this problem using exclusively the 3D geometric information provided by LIDARs. In this paper, we propose a PointNet++ based architecture to detect pedestrians in dense 3D point clouds. The aim is to explore the potential contribution of geometric information alone in pedestrian detection systems. We also present a semi-automatic labeling system that transfers pedestrian and non-pedestrian labels from RGB images onto the 3D domain. The fact that our datasets have RGB registered with point clouds enables label transferring by back projection from 2D bounding boxes to point clouds, with only a light manual supervision to validate results. We train PointNet++ with the geometry of the resulting 3D labelled clusters. The evaluation confirms the effectiveness of the proposed method, yielding precision and recall values around 98%.