用分层GRU与条件查询槽预测足球中球的10类动作，提升罕见动作识别效果。
Hierarchical GRU with Input-Conditioned Slot Queries for Ball Action Anticipation
- 分层结构：局部Transformer提取片段特征，GRU融合时间上下文，输入条件槽解码目标。
- 在SoccerNet数据集上达17.91% mAP，显著提升稀有动作预测性能。
- 适合关注动作预测、视频理解与稀有事件识别的研究者或应用开发者。
我们提出一种分层模型，用于足球广播视频中的球动作预测。给定30秒观察窗口，系统预测随后5秒内发生的10类动作。每个5秒子窗口内的片段特征由共享的局部Transformer编码；GRU在所有子窗口间聚合时间上下文；最终通过带K个输入条件事件槽的Transformer解码器，使用三个解耦头（存在性、类别、时间偏移）输出预测结果。引入频率重加权匈牙利匹配，系统性地增强对稀有动作类别的关注；同时采用高斯软标签进行时间区间监督。在SoccerNet Ball Action Anticipation基准上，该方法在测试服务器上取得17.91%的mAP。
原文摘要 · Abstract (English)
We present a hierarchical model for ball action anticipation in football broadcast video. Given a 30-second observation window, the system predicts actions occurring in the subsequent 5-second window across 10 classes. A shared local Transformer encodes clip-level features within each 5-second sub-window; a GRU then aggregates temporal context across all sub-windows; finally, a Transformer decoder with K input-conditioned event slots decodes the anticipation target via three decoupled heads (objectness, class, temporal offset). We introduce frequency-reweighted Hungarian matching that systematically favours rare action classes, and Gaussian soft targets for temporal bin supervision. On the SoccerNet Ball Action Anticipation benchmark, our method achieves 17.91% mAP on the test server.
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。