Send Less, Perceive More: Masked Quantized Point Cloud Communication for Loss-Tolerant Collaborative Perception

TL;DR

This paper presents QPoint2Comm, a novel quantized point-cloud communication framework for collaborative perception in vehicles, achieving high accuracy, low bandwidth usage, and robustness to packet loss through innovative encoding and training strategies.

Contribution

The paper introduces a new quantized point-cloud communication method with a shared codebook and masked training, improving bandwidth efficiency and robustness in collaborative perception.

Findings

Outperforms existing methods in accuracy and bandwidth efficiency

Maintains high performance under severe packet loss conditions

Introduces a cascade attention fusion module for better multi-vehicle data integration

Abstract

Collaborative perception allows connected vehicles to overcome occlusions and limited viewpoints by sharing sensory information. However, existing approaches struggle to achieve high accuracy under strict bandwidth constraints and remain highly vulnerable to random transmission packet loss. We introduce QPoint2Comm, a quantized point-cloud communication framework that dramatically reduces bandwidth while preserving high-fidelity 3D information. Instead of transmitting intermediate features, QPoint2Comm directly communicates quantized point-cloud indices using a shared codebook, enabling efficient reconstruction with lower bandwidth than feature-based methods. To ensure robustness to possible communication packet loss, we employ a masked training strategy that simulates random packet loss, allowing the model to maintain strong performance even under severe transmission failures. In addition, a cascade attention fusion module is proposed to enhance multi-vehicle information integration. Extensive experiments on both simulated and real-world datasets demonstrate that QPoint2Comm sets a new state of the art in accuracy, communication efficiency, and resilience to packet loss.