Towards Efficient 3D Object Detection for Vehicle-Infrastructure Collaboration via Risk-Intent Selection

TL;DR

This paper introduces RiSe, a risk-aware framework for vehicle-infrastructure perception that selectively transmits critical features, significantly reducing bandwidth while maintaining high detection accuracy in autonomous driving.

Contribution

The paper proposes a novel risk-intent selection approach with a potential field model and intention-driven prediction to improve communication efficiency in vehicle-infrastructure perception.

Findings

Reduces communication volume to 0.71% of full feature sharing.

Maintains state-of-the-art detection accuracy.

Establishes a Pareto optimal trade-off between bandwidth and perception performance.

Abstract

Vehicle-Infrastructure Collaborative Perception (VICP) is pivotal for resolving occlusion in autonomous driving, yet the trade-off between communication bandwidth and feature redundancy remains a critical bottleneck. While intermediate fusion mitigates data volume compared to raw sharing, existing frameworks typically rely on spatial compression or static confidence maps, which inefficiently transmit spatially redundant features from non-critical background regions. To address this, we propose Risk-intent Selective detection (RiSe), an interaction-aware framework that shifts the paradigm from identifying visible regions to prioritizing risk-critical ones. Specifically, we introduce a Potential Field-Trajectory Correlation Model (PTCM) grounded in potential field theory to quantitatively assess kinematic risks. Complementing this, an Intention-Driven Area Prediction Module (IDAPM) leverages ego-motion priors to proactively predict and filter key Bird's-Eye-View (BEV) areas essential for decision-making. By integrating these components, RiSe implements a semantic-selective fusion scheme that transmits high-fidelity features only from high-interaction regions, effectively acting as a feature denoiser. Extensive experiments on the DeepAccident dataset demonstrate that our method reduces communication volume to 0.71\% of full feature sharing while maintaining state-of-the-art detection accuracy, establishing a competitive Pareto frontier between bandwidth efficiency and perception performance.