Cyril666/VL3-Syn7M
收藏Hugging Face2025-02-07 更新2025-02-15 收录
下载链接：
https://hf-mirror.com/datasets/Cyril666/VL3-Syn7M
下载链接
链接失效反馈官方服务：
资源简介：
re-caption数据集是VideoLLaMA3训练期间使用的重标注数据，包含700万张多样化的高质量图片，每张图片都配有一个简短描述和一个详细描述。图片来源于COYO-700M、MS-COCO 2017、CC-3M和LLaVA-Pretrain数据集，并使用InternVL2进行了重标注。
The re-caption dataset is the re-annotated data used during the training of VideoLLaMA3, containing 7 million diverse, high-quality images, each accompanied by a short caption and a detailed caption. The images are sourced from COYO-700M, MS-COCO 2017, CC-3M, and LLaVA-Pretrain datasets, and re-annotated using InternVL2.
提供机构：
Cyril666
