Publication
Uncertainty-aware pseudo labels for domain adaptation in pedestrian trajectory prediction
Atanas Poibrenski; Farzad Nozarian; Farzaneh Rezaeianaran; Christian Müller
In: IEEE ITSC-2023. IEEE Intelligent Transportation Systems Conference (IEEE ITSC-2023), September 24-28, Bilbao, Spain, IEEExplore, 2023.
Abstract
Learning-based trajectory prediction models are
increasingly being used in a wide range of AI applications
such as autonomous driving. However, existing methods usually
ignore the potential distribution shift between the train and test
environments. This inevitably results in an increased prediction
error in the new domain. Towards this end, we present a
novel model-agnostic student-teacher model that leverages the
recent advances in self-training and utilizes predicted pseudo
trajectories from the target domain in order to improve its
generalization capabilities. More specifically, we propose to train
the model using both trajectories from the source domain and
predicted pseudo trajectories from the target domain. Since the
predicted trajectories can be noisy, we weight them by the
epistemic uncertainty of the model using MC-dropout, giving
more weight to the more certain ones. Additionally, we show
that the domain gap can be reduced further by augmenting
the source data. Experiments on the ETH/UCY datasets show
the effectiveness of our framework on domain adaptation for
pedestrian trajectory prediction.