Communication-Efficient and Context-Adaptive Collaborative Perception
Wenyu Lu, Hui Zhang, Yuquan Yang, ZiYin Zhang, Xiaohua Xu
AI summary
Problem
Existing collaborative perception methods apply uniform communication strategies across the entire field of view, causing inefficient bandwidth utilization and a poor trade-off between detection accuracy and transmission costs.
Approach
CaCP partitions the field of view into four contextual regions and adaptively selects between intermediate fusion, late fusion, or compact visibility masks for each area to prioritize bandwidth for high-value information.
Key results
- Achieves state-of-the-art accuracy-bandwidth trade-off on simulated and real-world datasets
- Reduces bandwidth consumption by up to 17% compared to prior methods
- Introduces a High-Precision Bounding Box Filter for reliable localization quality assessment
- Demonstrates effective hybrid fusion of intermediate and late strategies via spatial context adaptation
Why it matters
It enables practical, large-scale deployment of autonomous driving systems by overcoming critical V2X bandwidth limitations without sacrificing detection accuracy.
Abstract
Collaborative perception is pivotal for the large- scale deployment of autonomous driving, yet it has long grappled with the trade-off between perception accuracy and bandwidth consumption. Existing methods fail to analyze the fine-grained characteristics of Field of View (FoV), leading to inefficient bandwidth utilization. To address this, we propose a Context-adaptive Collaborative Perception framework, termed CaCP. This method optimizes bandwidth usage by employing distinct collaboration strategies for FoV under varying contexts, thereby reducing communication overhead while maintaining perception accuracy. Additionally, CaCP introduces a novel spatial fusion of intermediate and late fusion strategies, yielding a more flexible collaborative scheme. Extensive experiments across multiple datasets encompassing both simulated (OPV2V) and real-world (V2V4Real) scenarios demonstrate that CaCP establishes a new state-of-the-art trade-off between accuracy and bandwidth. Notably, it reduces bandwidth consumption by up to 17% compared to previous works while achieving competitive or superior perception performance.