A million-scale text-to-video prompt-gallery dataset

[WangWenhao0716]

Hi, We contribute the first dataset featuring 1.67 million unique text-to-video prompts and 6.69 million videos generated from 4 different state-of-the-art diffusion models. We hope it can help your Open-Sora plan.

Title：VidProM: A Million-scale Real Prompt-Gallery Dataset for Text-to-Video Diffusion Models

Arxiv：https://arxiv.org/abs/2403.06098

Project：https://github.com/WangWenhao0716/VidProM

Download：https://huggingface.co/datasets/WenhaoWang/VidProM

[LinB203]

Thanks for the heads up, we'll take it under advisement.

[chg0901]

@WangWenhao0716 could you please show us some samples for checking on your project link for further researching?

By the way, welcome to update your work in the MiniSora Dataset Section (https://github.com/mini-sora/minisora/tree/main#dataset)

[WangWenhao0716]

Yeah, thanks for your interest. I will upload an example folder with 10000 random prompts and corresponding videos. I am pleasure to update my work in the MiniSora Dataset Section :)

[WangWenhao0716]

@WangWenhao0716 could you please show us some samples for checking on your project link for further researching?

By the way, welcome to update your work in the MiniSora Dataset Section (https://github.com/mini-sora/minisora/tree/main#dataset)

I see it has been updated, thanks!

[WangWenhao0716]

@chg0901 Done: https://huggingface.co/datasets/WenhaoWang/VidProM/tree/main/example