HoopTransformer: Advancing NBA Offensive Play Recognition with Self-Supervised Learning from Player Trajectories,Sports Medicine

当前位置： X-MOL 学术 › Sports Med. › 论文详情

Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)

HoopTransformer: Advancing NBA Offensive Play Recognition with Self-Supervised Learning from Player Trajectories
Sports Medicine ( IF 9.3 ) Pub Date : 2024-05-30 , DOI: 10.1007/s40279-024-02030-3
Xing Wang ₁ , Zitian Tang _{2,

3} , Jianchong Shao ₂ , Sam Robertson ₄ , Miguel-Ángel Gómez ₁ , Shaoliang Zhang ₂

Affiliation

Background and Objective

Understanding and recognizing basketball offensive set plays, which involve intricate interactions between players, have always been regarded as challenging tasks for untrained humans, not to mention machines. In this study, our objective is to propose an artificial intelligence model that can automatically recognize offensive plays using a novel self-supervised learning approach.

Methods

The dataset was collected by SportVU from 632 games during the 2015–2016 season of the National Basketball Association (NBA), with a total of 90,524 possessions. A multi-agent motion prediction pretraining model was built on the basis of axial-attention transformer and trained with different masking strategies: motion prediction (MP), motion reconstruction (MR), and MP + MR joint strategy. A downstream play-level classification task and similarity search were used to evaluate the models’ performance.

Results

The results showed that the MP + MR joint masking strategy maximized the ability of the model compared with individual masking strategies. For the classification task, the joint strategy achieved a top-1 accuracy of 81.5% and top-3 accuracy of 97.5%. In the similarity search evaluation, the joint strategy attained a top-5 accuracy of 76% and top-10 accuracy of 59%. Additionally, with the same MP + MR joint masking strategy, our HoopTransformer model outperformed the two baseline models in the classification task and similarity search.

Conclusion

This study presents a self-supervised learning model and demonstrates the effectiveness and potential of the model in accurately comprehending and capturing player movements and complex interactions during offensive plays.

中文翻译：

HoopTransformer：通过对球员轨迹进行自我监督学习，推进 NBA 进攻识别

背景和目标

理解和识别篮球进攻定位球，其中涉及球员之间错综复杂的互动，一直被视为未经训练的人类具有挑战性的任务，更不用说机器了。在这项研究中，我们的目标是提出一种人工智能模型，该模型可以使用一种新颖的自我监督学习方法自动识别进攻性比赛。

方法

该数据集由 SportVU 从美国国家篮球协会（NBA） 2015-2016 赛季的 632 场比赛中收集，总共有 90,524 次控球。在轴向注意力转换器的基础上构建了多智能体运动预测预训练模型，并使用不同的掩码策略进行训练：运动预测（MP）、运动重建（MR）和 MP + MR 联合策略。使用下游游戏级别分类任务和相似性搜索来评估模型的性能。

结果

结果表明，与单个掩蔽策略相比，MP + MR 联合掩蔽策略使模型的能力最大化。对于分类任务，联合策略实现了 81.5% 的 top-1 准确率和 97.5% 的 top-3 准确率。在相似性搜索评估中，联合策略获得了 76% 的前 5 准确率和 59% 的前 10 准确率。此外，使用相同的 MP + MR 联合掩蔽策略，我们的 HoopTransformer 模型在分类任务和相似性搜索方面优于两个基线模型。