finetrainers/OpenVid-1k-split
收藏Hugging Face2025-04-10 更新2025-04-12 收录
下载链接：
https://hf-mirror.com/datasets/finetrainers/OpenVid-1k-split
下载链接
链接失效反馈官方服务：
资源简介：
OpenVid-1M是一个视频数据集，包含了1000个视频样本，每个样本都带有视频文件、文本描述、美学分数、运动分数、时间一致性分数等特征信息。数据集用于视频生成模型和文本到视频扩散模型的训练，样本通过筛选原始数据集中的视频得到，确保了视频质量。
OpenVid-1M is a video dataset consisting of 1000 video samples, each with a video file, text description, aesthetic score, motion score, temporal consistency score, and other feature information. The dataset is used for training video generative models and text-to-video diffusion models. The samples are obtained by filtering videos from the original dataset to ensure video quality.
提供机构：
finetrainers
