arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2603.06544cs.CV
收藏

多源多模态数据中冗余建模与测量用于自动驾驶

Modeling and Measuring Redundancy in Multisource Multimodal Data for Autonomous Driving

Dept. of Information Science University of North Texas Denton, Texas, USA(信息科学系 诺克斯维尔大学 德顿, 德克萨斯州, 美国);Dept. of Data Science University of North Texas Denton, Texas, USA(数据科学系 诺克斯维尔大学 德顿, 德克萨斯州, 美国)

Yuhan Zhou, Mehri Sattari, Haihua Chen, Kewei Sha

AI总结:

本文研究了自动驾驶中多源多模态数据的冗余问题,通过实验展示了删除冗余标签对目标检测性能的提升作用,并揭示了数据质量与性能之间的直接联系。

摘要:

下一代自动驾驶车辆(AVs)依赖大量多源和多模态(M²)数据来支持实时决策。在实践中,由于环境条件和传感器限制,数据质量(DQ)在不同来源和模态中有所不同,但AV研究主要侧重于算法设计而非DQ分析。本文聚焦于冗余作为自动驾驶数据集中的基本但未被充分探索的DQ问题。使用nuScenes和Argoverse 2(AV2)数据集,我们对多源摄像机数据和多模态图像-激光雷达数据进行建模和测量,并评估删除冗余标签对YOLOv8目标检测任务的影响。实验结果表明,选择性地从具有共享视野的摄像机中删除冗余图像目标标签可以提高检测效果。在nuScenes中,三个代表性重叠区域的mAP50从0.66到0.70,从0.64到0.67,以及从0.53到0.55,而在其他重叠摄像机对上的检测仍保持在基线水平,即使在更强的剪枝下也是如此。在AV2中,移除了4.1-8.6%的标签,mAP50仍接近0.64的基线。多模态分析也揭示了图像和激光雷达数据之间显著的冗余。这些发现表明,冗余是一个可测量且可操作的DQ因素,对AV性能有直接影响。本文强调冗余作为自动驾驶感知中的数据质量因素,并推动以数据为中心的视角来评估和改进AV数据集。代码、数据和实现细节可在https://github.com/yhZHOU515/RedundancyAD上公开获取。

英文摘要:

Next-generation autonomous vehicles (AVs) rely on large volumes of multisource and multimodal ($M^2$) data to support real-time decision-making. In practice, data quality (DQ) varies across sources and modalities due to environmental conditions and sensor limitations, yet AV research has largely prioritized algorithm design over DQ analysis. This work focuses on redundancy as a fundamental but underexplored DQ issue in AV datasets. Using the nuScenes and Argoverse 2 (AV2) datasets, we model and measure redundancy in multisource camera data and multimodal image-LiDAR data, and evaluate how removing redundant labels affects the YOLOv8 object detection task. Experimental results show that selectively removing redundant multisource image object labels from cameras with shared fields of view improves detection. In nuScenes, mAP${50}$ gains from $0.66$ to $0.70$, $0.64$ to $0.67$, and from $0.53$ to $0.55$, on three representative overlap regions, while detection on other overlapping camera pairs remains at the baseline even under stronger pruning. In AV2, $4.1$-$8.6\%$ of labels are removed, and mAP${50}$ stays near the $0.64$ baseline. Multimodal analysis also reveals substantial redundancy between image and LiDAR data. These findings demonstrate that redundancy is a measurable and actionable DQ factor with direct implications for AV performance. This work highlights the role of redundancy as a data quality factor in AV perception and motivates a data-centric perspective for evaluating and improving AV datasets. Code, data, and implementation details are publicly available at: https://github.com/yhZHOU515/RedundancyAD

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑