UniTraj: A Unified Framework for Scalable Vehicle Trajectory Prediction
This work addresses scalability issues in vehicle trajectory prediction for autonomous driving, though it is incremental as it builds on existing data-driven methods by providing a framework for better evaluation and insights.
The paper tackles the challenge of scaling vehicle trajectory prediction models across different datasets by introducing UniTraj, a unified framework that integrates datasets, models, and evaluations, finding that model performance drops during transfer but improves with larger, more diverse data, achieving a new state-of-the-art result on the nuScenes dataset.
Vehicle trajectory prediction has increasingly relied on data-driven solutions, but their ability to scale to different data domains and the impact of larger dataset sizes on their generalization remain under-explored. While these questions can be studied by employing multiple datasets, it is challenging due to several discrepancies, e.g., in data formats, map resolution, and semantic annotation types. To address these challenges, we introduce UniTraj, a comprehensive framework that unifies various datasets, models, and evaluation criteria, presenting new opportunities for the vehicle trajectory prediction field. In particular, using UniTraj, we conduct extensive experiments and find that model performance significantly drops when transferred to other datasets. However, enlarging data size and diversity can substantially improve performance, leading to a new state-of-the-art result for the nuScenes dataset. We provide insights into dataset characteristics to explain these findings. The code can be found here: https://github.com/vita-epfl/UniTraj