Generating Synthetic Ground Truth Distributions for Multi-step Trajectory Prediction using Probabilistic Composite B\'ezier Curves

arXiv:2404.04397v1 Announce Type: new
Abstract: An appropriate data basis grants one of the most important aspects for training and evaluating probabilistic trajectory prediction models based on neural networks. In this regard, a common shortcoming of current benchmark datasets is their limitation to sets of sample trajectories and a lack of actual ground truth distributions, which prevents the use of more expressive error metrics, such as the Wasserstein distance for model evaluation. Towards this end, this paper proposes a novel approach …

abstract arxiv benchmark cs.lg current data datasets grants networks neural networks prediction prediction models regard sample synthetic training trajectory truth type