Joint object detection and re-identification for 3D obstacle multi-camera systems

TL;DR

This paper presents a novel multi-task network for joint object detection and re-identification in 3D obstacle detection systems, improving accuracy and object tracking across multiple cameras in autonomous driving.

Contribution

It introduces a new network architecture that combines detection and re-identification, enhancing 3D detection quality and object tracking in multi-camera autonomous systems.

Findings

Over 5% improvement in car detection accuracy in overlapping camera areas

Effective integration of re-identification with detection enhances multi-camera tracking

Outperforms traditional NMS techniques in experimental evaluations

Abstract

In recent years, the field of autonomous driving has witnessed remarkable advancements, driven by the integration of a multitude of sensors, including cameras and LiDAR systems, in different prototypes. However, with the proliferation of sensor data comes the pressing need for more sophisticated information processing techniques. This research paper introduces a novel modification to an object detection network that uses camera and lidar information, incorporating an additional branch designed for the task of re-identifying objects across adjacent cameras within the same vehicle while elevating the quality of the baseline 3D object detection outcomes. The proposed methodology employs a two-step detection pipeline: initially, an object detection network is employed, followed by a 3D box estimator that operates on the filtered point cloud generated from the network's detections. Extensive experimental evaluations encompassing both 2D and 3D domains validate the effectiveness of the proposed approach and the results underscore the superiority of this method over traditional Non-Maximum Suppression (NMS) techniques, with an improvement of more than 5\% in the car category in the overlapping areas.