论文标题
轨迹预测中跨数据库性能的不确定性估计
Uncertainty estimation for Cross-dataset performance in Trajectory prediction
论文作者
论文摘要
尽管已经在开发轨迹预测方法方面进行了许多工作，并且已经提出了各种数据集来基准这项任务，但到目前为止，很少对这些方法在跨数据集的通用性和可传递性进行研究。在本文中，我们观察到四个不同数据集（Argoverse，Nuscenes，互动，移位）的两种最新最新轨迹预测方法的性能。该分析允许对最新轨迹预测模型的概括性提供一些见解，并分析哪个数据集更代表真实的驾驶场景，因此可以更好地传递性能。此外，我们提出了一种新的方法来估计预测不确定性，并显示如何使用它来在数据集中取得更好的性能。
While a lot of work has been carried on developing trajectory prediction methods, and various datasets have been proposed for benchmarking this task, little study has been done so far on the generalizability and the transferability of these methods across dataset. In this paper, we observe the performance of two of the latest state-of-the-art trajectory prediction methods across four different datasets (Argoverse, NuScenes, Interaction, Shifts). This analysis allows to gain some insights on the generalizability proprieties of most recent trajectory prediction models and to analyze which dataset is more representative of real driving scenes and therefore enables better transferability. Furthermore we present a novel method to estimate prediction uncertainty and show how it could be used to achieve better performance across datasets.
