MatteoGinesi/violence-detection-vivit_32frame
收藏Hugging Face2025-04-11 更新2025-04-12 收录
下载链接:
https://hf-mirror.com/datasets/MatteoGinesi/violence-detection-vivit_32frame
下载链接
链接失效反馈官方服务:
资源简介:
这是一个为暴力检测任务准备的预处理视频剪辑数据集,包含224x224像素大小的32帧剪辑,适用于训练ViViT变压器模型。数据集分为训练集和测试集,共有两个标签类别:非暴力和暴力。
This is a preprocessed video clip dataset for violence detection, containing 32-frame clips of 224x224 pixels size, prepared for training ViViT transformer models. The dataset is split into training and test sets, with two label categories: Non Violence and Violence.
提供机构:
MatteoGinesi



