TIP-I2V: A Million-Scale Real Text and Image Prompt Dataset for Image-to-Video Generation

TL;DR

TIP-I2V is a large-scale dataset of over 1.7 million user prompts for image-to-video generation, enabling research on prompt analysis, model evaluation, and safety improvements in this emerging field.

Contribution

The paper introduces TIP-I2V, the first extensive dataset of text and image prompts specifically for image-to-video generation, filling a critical gap in available resources.

Findings

Compared TIP-I2V with VidProM and DiffusionDB highlighting key differences.

Demonstrated potential uses of TIP-I2V for model analysis and safety enhancement.

Provided a large, curated dataset to advance image-to-video research.

Abstract

Video generation models are revolutionizing content creation, with image-to-video models drawing increasing attention due to their enhanced controllability, visual consistency, and practical applications. However, despite their popularity, these models rely on user-provided text and image prompts, and there is currently no dedicated dataset for studying these prompts. In this paper, we introduce TIP-I2V, the first large-scale dataset of over 1.70 million unique user-provided Text and Image Prompts specifically for Image-to-Video generation. Additionally, we provide the corresponding generated videos from five state-of-the-art image-to-video models. We begin by outlining the time-consuming and costly process of curating this large-scale dataset. Next, we compare TIP-I2V to two popular prompt datasets, VidProM (text-to-video) and DiffusionDB (text-to-image), highlighting differences in both basic and semantic information. This dataset enables advancements in image-to-video research. For instance, to develop better models, researchers can use the prompts in TIP-I2V to analyze user preferences and evaluate the multi-dimensional performance of their trained models; and to enhance model safety, they may focus on addressing the misinformation issue caused by image-to-video models. The new research inspired by TIP-I2V and the differences with existing datasets emphasize the importance of a specialized image-to-video prompt dataset. The project is available at https://tip-i2v.github.io.