MatteoGinesi/violence-detection-vivit

这是一个用于暴力检测的预处理视频片段数据集，适用于训练ViViT变换器模型，包含10帧的片段。数据集标签分为两类：0代表非暴力，1代表暴力。原始数据集来源于Mohamed Mustafa的Real Life Violence Situations Dataset。视频帧大小为224x224，每两帧采样一帧，共10帧组成一个片段。数据集分为训练集和测试集。数据集遵循CC BY 4.0许可。This dataset contains preprocessed video clips (224x224) for violence detection, prepared for training ViViT transformer models with 10-frame clips. The labels are categorized into two types: 0 for Non Violence and 1 for Violence. The original dataset is sourced from Real Life Violence Situations Dataset by M. Elesawy, M. Hussein, and M. A. El Massih. The frame size is 224x224, with 10 frames sampled every 2 frames to form a clip. The dataset is split into training and test sets, and it is licensed under CC BY 4.0.