SNOW

SNOW数据集由滑铁卢大学系统设计工程系创建，专注于板球比赛中裁判姿态检测。该数据集包含五类裁判动作的图像，总计390张，来源于YouTube和Google图片。创建过程中，使用了预训练的卷积神经网络进行特征提取。SNOW数据集主要应用于自动生成板球比赛精彩片段，旨在通过识别裁判的特定手势和姿态来捕捉比赛中的关键事件。

The SNOW Dataset was developed by the Department of Systems Design Engineering at the University of Waterloo, focusing on referee pose detection in cricket matches. It contains 390 images in total covering five types of referee actions, sourced from YouTube and Google Images. During its creation, a pre-trained convolutional neural network was employed for feature extraction. The SNOW Dataset is primarily used for the automatic generation of cricket match highlight reels, with the goal of capturing key events in matches by recognizing specific referee gestures and poses.