A Late Collaborative Perception Framework for 3D Multi-Object and Multi-Source Association and Fusion

TL;DR

This paper introduces a late collaborative perception framework for 3D multi-object fusion in autonomous driving, achieving high accuracy without sharing proprietary detection models, thus addressing real-world communication and privacy constraints.

Contribution

The paper proposes a novel late fusion method that operates solely on shared 3D bounding box attributes, setting a new state-of-the-art in accuracy and efficiency for multi-agent perception.

Findings

Up to 5x lower position error compared to existing methods

Scale error reduced by a factor of 7.5

Orientation error halved while maintaining 100% precision and recall

Abstract

In autonomous driving, recent research has increasingly focused on collaborative perception based on deep learning to overcome the limitations of individual perception systems. Although these methods achieve high accuracy, they rely on high communication bandwidth and require unrestricted access to each agent's object detection model architecture and parameters. These constraints pose challenges real-world autonomous driving scenarios, where communication limitations and the need to safeguard proprietary models hinder practical implementation. To address this issue, we introduce a novel late collaborative framework for 3D multi-source and multi-object fusion, which operates solely on shared 3D bounding box attributes-category, size, position, and orientation-without necessitating direct access to detection models. Our framework establishes a new state-of-the-art in late fusion, achieving up to five times lower position error compared to existing methods. Additionally, it reduces scale error by a factor of 7.5 and orientation error by half, all while maintaining perfect 100% precision and recall when fusing detections from heterogeneous perception systems. These results highlight the effectiveness of our approach in addressing real-world collaborative perception challenges, setting a new benchmark for efficient and scalable multi-agent fusion.