[论文] How good was my shot? Quantifying Player Skill Level in Table Tennis

## 论文概要 **研究领域**: CV **作者**: Akihiro Kubota, Tomoya Hasegawa, Ryo Kawahara, Ko Nishino **发布时间**: 2026-03-26 **arXiv**: [2603.25736](https://arxiv.org/abs/2603.25736) ## 中文摘要 评估个人技能水平至关重要，因为它内在地塑造了人们的行为。然而，量化技能具有挑战性，因为它是潜藏于观察到的行为中的。为探索人类行为中的技能理解，我们聚焦于双人运动——特别是乒乓球——在这类运动中，技能不仅体现在复杂的动作中，更体现在基于游戏情境执行时的微妙差异中。我们的核心思想是学习每个球员战术性球拍击球的生成模型，并将它们联合嵌入到一个共同的潜在空间中，该空间编码个体特征，包括与技能水平相关的特征。通过在大量 3D 重建的专业比赛数据集上训练这些球员模型，并将它们条件化于全面的游戏情境——包括球员站位和对手行为——这些模型在其潜在空间内捕获了个体的战术特征。我们探查这个学习到的球员空间，发现它反映了不同的比赛风格和属性，这些共同代表了技能水平。通过在这些嵌入上训练一个简单的相对排名网络，我们证明了可以实现相对和绝对技能预测。这些结果表明，学习到的球员空间有效地量化了技能水平，为复杂交互行为中的自动化技能评估奠定了基础。 ## 原文摘要 Gauging an individual's skill level is crucial, as it inherently shapes their behavior. Quantifying skill, however, is challenging because it is latent to the observed actions. To explore skill understanding in human behavior, we focus on dyadic sports -- specifically table tennis -- where skill manifests not just in complex movements, but in the subtle nuances of execution conditioned on game context. Our key idea is to learn a generative model of each player's tactical racket strokes and jointly embed them in a common latent space that encodes individual characteristics, including those pertaining to skill levels. By training these player models on a large-scale dataset of 3D-reconstructed professional matches and conditioning them on comprehensive game context -- including player positi... --- *自动采集于 2026-03-28* #论文 #arXiv #CV #小凯