arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2203.01437cs.CV
收藏

MUAD:自动驾驶的多重不确定性,多种不确定性类型和任务的基准

MUAD: Multiple Uncertainties for Autonomous Driving, a benchmark for multiple uncertainty types and tasks

ENSTA Paris(巴黎高等先进技术学院);IP Paris(巴黎理工学院);Paris-Saclay University(巴黎萨克雷大学);Anyverse;Aix Marseille University(艾克斯-马赛大学)

Gianni Franchi, Xuanlong Yu, Andrei Bursuc, Angel Tena, Rémi Kazmierczak, Séverine Dubuisson, Emanuel Aldea, David Filliat

摘要:

预测性不确定性估计对于在现实世界的自治系统中安全部署深度神经网络至关重要。然而,对于大多数数据集来说,解开不同类型和来源的不确定性并不简单,特别是因为不确定性没有基础事实。此外,虽然不同强度的不利天气条件可能会破坏神经网络预测,但它们通常在公共数据集中的训练集和测试集中都表现不足。(自动驾驶的多重解锁),由10,413张具有各种恶劣天气条件的逼真合成图像组成(夜、雾、雨、雪)、分布外对象以及用于语义分割、深度估计、对象和实例检测的注释。MUAD允许更好地评估不同来源的不确定性对模型性能的影响。我们在多个任务中对几个基线深度神经网络的这种影响进行了彻底的实验研究,并发布了我们的数据集,以允许研究人员在不利条件下有条不紊地对他们的算法进行基准测试。更多的可视化和MUAD的下载链接可以在https://muad-dataset.github.io/上找到。

英文摘要:

Predictive uncertainty estimation is essential for safe deployment of Deep Neural Networks in real-world autonomous systems. However, disentangling the different types and sources of uncertainty is non trivial for most datasets, especially since there is no ground truth for uncertainty. In addition, while adverse weather conditions of varying intensities can disrupt neural network predictions, they are usually under-represented in both training and test sets in public datasets.We attempt to mitigate these setbacks and introduce the MUAD dataset (Multiple Uncertainties for Autonomous Driving), consisting of 10,413 realistic synthetic images with diverse adverse weather conditions (night, fog, rain, snow), out-of-distribution objects, and annotations for semantic segmentation, depth estimation, object, and instance detection. MUAD allows to better assess the impact of different sources of uncertainty on model performance. We conduct a thorough experimental study of this impact on several baseline Deep Neural Networks across multiple tasks, and release our dataset to allow researchers to benchmark their algorithm methodically in adverse conditions. More visualizations and the download link for MUAD are available at https://muad-dataset.github.io/.

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。

↑