WenhaoWang/TIP-I2V

TIP-I2V是第一个包含超过170万条用户提供的文本和图像提示的数据集。除了提示外，TIP-I2V还包括由五种最先进的图像到视频生成模型（Pika、Stable Video Diffusion、Open-Sora、I2VGen-XL和CogVideoX-5B）生成的视频。TIP-I2V旨在促进更好、更安全的图像到视频模型的发展。

The TIP-I2V dataset is a large-scale dataset for image-to-video generation tasks, comprising over 1.70 million unique user-provided text and image prompts. Besides the prompts, TIP-I2V also includes videos generated by five state-of-the-art image-to-video models (Pika, Stable Video Diffusion, Open-Sora, I2VGen-XL, and CogVideoX-5B). The dataset is designed to contribute to the development of better and safer image-to-video models. It includes various splits (Full, Subset, Eval) and provides options for downloading text and image prompts, embeddings, and generated videos. The dataset is licensed under the CC BY-NC 4.0 license.