重新审视基于视觉Transformer的危险擒抱检测
Revisiting Risky Tackle Detection with Vision Transformers
浏览论文内容
中文总结 AI 辅助
本文复现了ICPR 2026中基于ViViT的危险擒抱检测研究，确认run_15的召回率0.667和F1 0.588与原文一致，并发现亮度增强是主导因素，无增强时模型不及C3D基线。
中文摘要 AI 辅助
本文是ICPR 2026关于美式橄榄球训练视频中危险擒抱检测研究的一项Track 2可复现性配套工作。原始研究在733个标注有SATT-3评分标准的片段上微调了视频视觉Transformer（ViViT），使用了焦点损失、田口L18增强和5折交叉验证，并报告了危险类别的召回率为0.67，危险类别的F1分数为0.59。本配套工作记录了发布的工件，并将这些数字追溯到具体的脚本、折输出和聚合文件。复现的主要结果是run_15，它结合了高斯噪声与静态亮度降低，不使用旋转和翻转。其折平均危险召回率为0.667，折平均危险F1分数为0.588，这些值在四舍五入后与已发布的主要结果一致。消融实验表明，亮度是主导因素，其危险召回率的主效应范围为0.055，大于旋转、翻转和噪声的范围。在没有增强的情况下，ViViT达到的危险召回率为0.545，未超过C3D基线的0.583。原始片段显示了可识别的学生运动员，因此不能重新分发。该工件为流程检查提供了公开样本，并为完整数据审查提供了受控途径。
英文摘要
This paper is a Track 2 reproducibility companion to an ICPR 2026 study on risky tackle detection in American football prac- tice videos. The original work fine-tuned a Video Vision Transformer (ViViT) on 733 clips labeled with the SATT-3 rubric. It used focal loss, Taguchi L18 augmentation, and 5-fold cross-validation. It reported risky- class recall of 0.67 and risky-class F1 of 0.59. This companion documents the released artifact and traces those numbers to specific scripts, fold out- puts, and aggregation files. The reproduced headline is run_15. It com- bines Gaussian noise with static brightness decrease and uses no rotation and no flip. Its fold-mean risky recall is 0.667 and its fold-mean risky F1 is 0.588. These values match the published headline after rounding. The ablation shows that brightness is the dominant factor. Its risky-recall main-effect range is 0.055, which is larger than the ranges for rotation, flip, and noise. Without augmentation, ViViT reaches risky recall of 0.545 and does not exceed the C3D baseline of 0.583. The raw clips show iden- tifiable student athletes, so they cannot be redistributed. The artifact provides a public sample for pipeline checks and a controlled route for full-data review.
发表机构
- Kansas State University（堪萨斯州立大学）
- Albright College（奥尔布赖特学院）
机构由 AI 辅助整理，请以论文原文为准。
