基于玩家注意力与共识融合，提升足球比赛球动事件检测准确率
SoccerNet 2026 Player-Centric Ball Action Spotting: Per-Player Attention with Agreement-Based Ensembling
- 分阶段设计：先生成球员动作置信度，再用时空注意力建模事件序列
- 引入空间优先注意力机制，使验证集宏平均F1提升1.87%
- 多模型加权融合并过滤异常误报，特别优化罕见铲球事件
我们提交了SoccerNet 2026球员中心球动事件定位挑战赛的方案，采用两阶段流程：首先通过轨迹感知动作检测器（TAAD）从转播视频中生成每个球员的动作置信度；其次使用去噪序列变换器（DST）Transformer将比赛状态特征与TAAD输出转化为结构化事件序列。我们改进了TAAD，引入时间注意力以增强跨帧上下文信息，并修复多个训练问题。在DST阶段，提出一种双阶段球员注意力机制，作用于比赛状态特征，发现空间优先注意力顺序（先跨球员后时间）使验证集宏平均F1提升1.87%。为利用架构多样性，训练四个模型变体，并采用加权事件融合方法，通过共识过滤抑制单模型误报，同时保持召回率，且对罕见铲球类事件设专门处理机制。最终系统将基准宏平均F1从48.6提升至58.94。
原文摘要 · Abstract (English)
We present our submission to the SoccerNet 2026 Player-Centric Ball Action Spotting challenge, which uses a two-stage pipeline: a Track-Aware Action Detector (TAAD) produces per-player action logits from broadcast video, and a Denoising Sequence Transduction (DST) transformer converts game-state features and TAAD logits into structured event sequences. We improve the TAAD with a temporal transformer that adds cross-frame context, alongside several training fixes. For the DST stage, we introduce a two-stage per-player attention mechanism operating on game-state features, and show that a spatial-first attention ordering (cross-player attention before temporal attention) improves validation Macro-F1 by 1.87%. To exploit architectural diversity, we train four model variants and combine them with a Weighted Event Fusion ensemble that applies agreement filtering to suppress single-model false positives while preserving recall, plus a dedicated exception for the rare tackle class. Our final system improves the challenge Macro-F1 from a baseline of 48.6 to 58.94.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。