ADAPT: Efficient Multi-Agent Trajectory Prediction with Adaptation

TL;DR

ADAPT is a novel multi-agent trajectory prediction method that uses dynamic weight learning to improve accuracy and efficiency in complex traffic scenes, outperforming existing methods with less computational cost.

Contribution

It introduces a dynamic weight learning approach with adaptive heads and endpoint-conditioned prediction, enhancing multi-agent trajectory forecasting without increasing model size.

Findings

Outperforms state-of-the-art methods on Argoverse and Interaction datasets.

Achieves higher accuracy with reduced computational overhead.

Utilizes adaptive prediction focusing on individual agents effectively.

Abstract

Forecasting future trajectories of agents in complex traffic scenes requires reliable and efficient predictions for all agents in the scene. However, existing methods for trajectory prediction are either inefficient or sacrifice accuracy. To address this challenge, we propose ADAPT, a novel approach for jointly predicting the trajectories of all agents in the scene with dynamic weight learning. Our approach outperforms state-of-the-art methods in both single-agent and multi-agent settings on the Argoverse and Interaction datasets, with a fraction of their computational overhead. We attribute the improvement in our performance: first, to the adaptive head augmenting the model capacity without increasing the model size; second, to our design choices in the endpoint-conditioned prediction, reinforced by gradient stopping. Our analyses show that ADAPT can focus on each agent with adaptive prediction, allowing for accurate predictions efficiently. https://KUIS-AI.github.io/adapt