We present RefPtsFusion, a lightweight and interpretable framework for cooperative autonomous driving. Instead of sharing large feature maps or query embeddings, vehicles exchange compact reference points, e.g., objects' positions, velocities, and size information. This approach shifts the focus from "what is seen" to "where to see", creating a sensor- and model-independent interface that works well across vehicles with heterogeneous perception models while greatly reducing communication bandwidth. To enhance the richness of shared information, we further develop a selective Top-K query fusion that selectively adds high-confidence queries from the sender. It thus achieves a strong balance between accuracy and communication cost. Experiments on the M3CAD dataset show that RefPtsFusion maintains stable perception performance while reducing communication overhead by five orders of magnitude, dropping from hundreds of MB/s to only a few KB/s at 5 FPS (frame per second), compared to traditional feature-level fusion methods. Extensive experiments also demonstrate RefPtsFusion's strong robustness and consistent transmission behavior, highlighting its potential for scalable, real-time cooperative driving systems.

翻译：本文提出RefPtsFusion，一种面向协同自动驾驶的轻量化可解释框架。车辆间不共享庞大的特征图或查询嵌入，而是交换紧凑的参考点，例如目标的位置、速度和尺寸信息。该方法将关注点从“看到什么”转向“看向何处”，创建了一个与传感器及模型无关的接口，能在搭载异构感知模型的车辆间高效工作，同时大幅降低通信带宽。为增强共享信息的丰富度，我们进一步开发了选择性Top-K查询融合机制，有选择地添加发送方的高置信度查询。由此在精度与通信成本之间实现了良好平衡。在M3CAD数据集上的实验表明，与传统特征级融合方法相比，RefPtsFusion在保持稳定感知性能的同时，将通信开销降低了五个数量级——在5 FPS（帧每秒）下从数百MB/s降至仅数KB/s。大量实验还证明了RefPtsFusion具有强鲁棒性和稳定的传输特性，凸显了其在可扩展实时协同驾驶系统中的应用潜力。