Faster-HEAL: An Efficient and Privacy-Preserving Collaborative Perception Framework for Heterogeneous Autonomous Vehicles
This work provides a practical and efficient solution for scalable collaborative perception in real-world scenarios involving heterogeneous autonomous vehicles, which is an incremental improvement over existing methods.
The paper addresses the challenge of heterogeneous autonomous vehicles in collaborative perception, where diverse sensors and models create a feature domain gap. Faster-HEAL proposes a lightweight framework that fine-tunes a low-rank visual prompt to align features, reducing trainable parameters by 94% and improving detection performance by 2% on the OPV2V-H dataset.
Collaborative perception (CP) is a promising paradigm for improving situational awareness in autonomous vehicles by overcoming the limitations of single-agent perception. However, most existing approaches assume homogeneous agents, which restricts their applicability in real-world scenarios where vehicles use diverse sensors and perception models. This heterogeneity introduces a feature domain gap that degrades detection performance. Prior works address this issue by retraining entire models/major components, or using feature interpreters for each new agent type, which is computationally expensive, compromises privacy, and may reduce single-agent accuracy. We propose Faster-HEAL, a lightweight and privacy-preserving CP framework that fine-tunes a low-rank visual prompt to align heterogeneous features with a unified feature space while leveraging pyramid fusion for robust feature aggregation. This approach reduces the trainable parameters by 94%, enabling efficient adaptation to new agents without retraining large models. Experiments on the OPV2V-H dataset show that Faster-HEAL improves detection performance by 2% over state-of-the-art methods with significantly lower computational overhead, offering a practical solution for scalable heterogeneous CP.