UECP: Uncertainty-Enhanced Collaborative Perception
2026-06-22 • Computer Vision and Pattern Recognition
Computer Vision and Pattern Recognition
AI summaryⓘ
The authors address improving how self-driving cars share and combine what they see to make better decisions. They point out that previous methods used confidence maps tied to detection results, which can be biased. Instead, the authors propose an uncertainty map based on real sensor data, like LiDAR point density, to provide clearer and more reliable evidence of perception quality. They build a fusion system using this map to better combine information from multiple cars, making the overall perception more accurate and less noisy. Their experiments show this method works better than existing approaches.
Collaborative perceptionAutonomous drivingUncertainty mapLiDARSensor fusionConfidence mapFeature fusionResidual fusionDetection headPoint density
Authors
Kang Yang, Tianci Bu, Peng Wang, Deying Li, Wen Jie, Yongcai Wang
Abstract
Collaborative perception serves as a pivotal solution to enhance the perception capability of individual agents in autonomous driving, where a core challenge lies in seeking reliable evidence to quantify and weight the contribution of each participating agent. Existing methods typically rely on a confidence map, which is co-trained with the detection head, but it is inherently correlated with the detection results and thus fails to provide unbiased physical evidence. Furthermore, how to deeply integrate evidence into the cooperative fusion process remains an open question. To address these issues, this paper first proposes an uncertainty map, a physically grounded and unambiguous metric for evaluating perception quality. This map is directly supervised by real-time sensor signals, i.e., LiDAR point density, ensuring decoupling from detection noise and thereby providing physical scenario-aware evidence for weighting agent contribution. Based on this map, we develop the Uncertainty-Enhanced Collaborative Perception (UECP) framework, centered on the Uncertainty-Aware Pyramid Fusion (UAPF) module. UAPF uses a coarse-to-fine strategy, with two key components: Uncertainty-Weighted Downsampling (UWD) for high-fidelity feature preservation, and Uncertainty-Guided Residual Fusion (UGRF) to reinforce ego features, suppressing noise and ensuring robust fusion. Extensive experiments on real-world datasets show UECP outperforms state-of-the-art methods in effectiveness and robustness by embedding the uncertainty map into fusion. Code will be publicly available.