zigg-ai/content-regions-distrib-yolo-23k
收藏Hugging Face2025-10-22 更新2025-10-25 收录
下载链接：
https://hf-mirror.com/datasets/zigg-ai/content-regions-distrib-yolo-23k
下载链接
链接失效反馈官方服务：
资源简介：
本数据集包含7种类别的YOLO格式内容标注：数字图形、数字演示、数字屏幕录制、数字视频游戏、多发言人视频片段、视频场景和视频谈话头部。数据集结构包括训练集和验证集，共有18325张训练图像和4582张验证图像。数据集目录包含图像和标签文件夹，其中标签文件夹中的每个文件包含以类别ID、中心点坐标、宽度和高度为格式的标注。
This dataset contains 7 classes of content annotations in YOLO format: digital graphics, digital presentation, digital screencast, digital videogame, footage with multiple speakers, footage of a scene, and footage of a talking head. The dataset structure includes a training set and a validation set, with a total of 18,325 training images and 4,582 validation images. The dataset directory contains folders for images and labels, where each file in the labels folder contains annotations in the format of class_id, x_center, y_center, width, and height.
提供机构：
zigg-ai
