TokenFlow: Consistent Diffusion Features for Consistent Video Editing
This addresses the challenge of user control and visual quality in video generation for AI applications, though it is incremental as it builds on existing diffusion models.
The authors tackled the problem of text-driven video editing by proposing a framework that uses a text-to-image diffusion model to generate high-quality videos that adhere to target text while preserving input video layout and motion, achieving state-of-the-art results on real-world videos.
The generative AI revolution has recently expanded to videos. Nevertheless, current state-of-the-art video models are still lagging behind image models in terms of visual quality and user control over the generated content. 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. Specifically, given a source video and a target text-prompt, our method generates a high-quality video that adheres to the target text, while preserving the spatial layout and motion of the input video. Our method is based on a key observation that consistency in the edited video can be obtained by enforcing consistency in the diffusion feature space. We achieve this by explicitly propagating diffusion features based on inter-frame correspondences, readily available in the model. Thus, our framework does not require any training or fine-tuning, and can work in conjunction with any off-the-shelf text-to-image editing method. We demonstrate state-of-the-art editing results on a variety of real-world videos. Webpage: https://diffusion-tokenflow.github.io/