A Graph Attention Based Approach for Trajectory Prediction in Multi-agent Sports Games
This work provides an incremental improvement in trajectory prediction for multi-agent sports games by better capturing agent dependencies.
This paper addresses multi-agent trajectory prediction in sports games, specifically basketball and soccer. The authors propose a graph attention-based model combined with a Temporal Convolutional Network (TCN) to capture fine-grained dependencies among coordinated agents, resulting in substantially improved performance compared to related approaches.
This work investigates the problem of multi-agents trajectory prediction. Prior approaches lack of capability of capturing fine-grained dependencies among coordinated agents. In this paper, we propose a spatial-temporal trajectory prediction approach that is able to learn the strategy of a team with multiple coordinated agents. In particular, we use graph-based attention model to learn the dependency of the agents. In addition, instead of utilizing the recurrent networks (e.g., VRNN, LSTM), our method uses a Temporal Convolutional Network (TCN) as the sequential model to support long effective history and provide important features such as parallelism and stable gradients. We demonstrate the validation and effectiveness of our approach on two different sports game datasets: basketball and soccer datasets. The result shows that compared to related approaches, our model that infers the dependency of players yields substantially improved performance. Code is available at https://github.com/iHeartGraph/predict