In autonomous driving, goal-based multi-trajectory prediction methods are proved to be effective recently, where they first score goal candidates, then select a final set of goals, and finally complete trajectories based on the selected goals. However, these methods usually involve goal predictions based on sparse predefined anchors. In this work, we propose an anchor-free model, named DenseTNT, which performs dense goal probability estimation for trajectory prediction. Our model achieves state-of-the-art performance, and ranks 1st on the Waymo Open Dataset Motion Prediction Challenge. Project page is at https://github.com/Tsinghua-MARS-Lab/DenseTNT.
@article{arxiv.2106.14160,
title = {DenseTNT: Waymo Open Dataset Motion Prediction Challenge 1st Place Solution},
author = {Junru Gu and Qiao Sun and Hang Zhao},
journal= {arXiv preprint arXiv:2106.14160},
year = {2021}
}
Comments
This is a technical report. ICCV paper is at arXiv:2108.09640 and project page is at https://github.com/Tsinghua-MARS-Lab/DenseTNT