In this work, we present a framework that harnesses the power of a text-to-image diffusion model for the task of text-driven video editing.

We use Stable Diffusion as our pre-trained text-to-image model; we use the StableDiffusion-v-2-1 checkpoint provided via official HuggingFace webpage.

We presented a new framework for text-driven video editing using an image diffusion model.

In this experiment, we replace TokenFlow with extended attention and compute it between each frames of the edited video and the keyframes (w joint attention).

Our method outputs videos that better adhere to the edit prompt while maintaining temporal consistency of the resulting edited video

Here's how to edit propagation via our framework, TokenFlow.

Take a peek at how TokenFlow works and the method we used.

Tired of hearing about Diffusion models and stable diffusion but not knowing what they mean? Well, you don't have to worry about that anymore!

With the evergrowing scale of vision-language datasets and pretraining strategies, there has been a remarkable progress in text-driven image generation