TokenFlow enables consistent video editing, without requiring any training or fine-tuning, and can work in conjunction with any off-the-shelf text-to-image editing method.
TokenFlow enables consistent video editing, using a pre-trained text-to-image diffusion model, without any further training or finetuning.
TokenFlow uses inter-frame correspondences readily available in the model to enforce consistency in the diffusion feature space.
TokenFlow can work in conjunction with any off-the-shelf text-to-image editing method.
TokenFlow does not require any training or fine-tuning, making it a convenient and efficient solution for video editing.
TokenFlow is licensed under the MIT license, making it freely available for use and modification.
Video editing for social media
Video editing for advertising
Video editing for film and television
Video editing for education and training
Install the TokenFlow framework
Prepare your video data
Use the TokenFlow framework to edit your video
Save and export your edited video