ParCon: Noise-Robust Collaborative Perception via Multi-module Parallel Connection

TL;DR

ParCon is a novel parallel-connection architecture for collaborative perception in autonomous vehicles, significantly improving noise robustness and detection accuracy while reducing computational costs.

Contribution

It introduces a multi-module parallel connection framework that enhances noise robustness and efficiency in collaborative perception systems.

Findings

Increases detection accuracy by 6.91% in noisy environments

Reduces FLOPs by 11.46%, improving computational efficiency

Demonstrates robustness to noise through parallel module design

Abstract

In this paper, we investigate improving the perception performance of autonomous vehicles through communication with other vehicles and road infrastructures. To this end, we introduce a novel collaborative perception architecture, called ParCon, which connects multiple modules in parallel, as opposed to the sequential connections used in most other collaborative perception methods. Through extensive experiments, we demonstrate that ParCon inherits the advantages of parallel connection. Specifically, ParCon is robust to noise, as the parallel architecture allows each module to manage noise independently and complement the limitations of other modules. As a result, ParCon achieves state-of-the-art accuracy, particularly in noisy environments, such as real-world datasets, increasing detection accuracy by 6.91%. Additionally, ParCon is computationally efficient, reducing floating-point operations (FLOPs) by 11.46%.