Vidu: a Highly Consistent, Dynamic and Skilled Text-to-Video Generator with Diffusion Models

TL;DR

Vidu is a state-of-the-art diffusion-based text-to-video generator capable of producing high-resolution, long, and coherent videos with diverse content, demonstrating advanced scalability and control features.

Contribution

Introducing Vidu, a novel diffusion model with U-ViT backbone that significantly improves scalability, coherence, and control in text-to-video generation tasks.

Findings

Produces 1080p videos up to 16 seconds long

Achieves strong coherence and dynamism in generated videos

Performs well on controllable video generation tasks

Abstract

We introduce Vidu, a high-performance text-to-video generator that is capable of producing 1080p videos up to 16 seconds in a single generation. Vidu is a diffusion model with U-ViT as its backbone, which unlocks the scalability and the capability for handling long videos. Vidu exhibits strong coherence and dynamism, and is capable of generating both realistic and imaginative videos, as well as understanding some professional photography techniques, on par with Sora -- the most powerful reported text-to-video generator. Finally, we perform initial experiments on other controllable video generation, including canny-to-video generation, video prediction and subject-driven generation, which demonstrate promising results.