ToonCrafter (512x320) Generative Cartoon Interpolation Model Card

ToonCrafter (512x320) is a video diffusion model that
takes in two still images as conditioning images and text prompt describing dynamics,
and generates interpolation videos from them.

Model Details

Model Description

ToonCrafter, a generative cartoon interpolation approach, aims to generate
short video clips (~2 seconds) from two conditioning images (starting frame and ending frame) and text prompt.

This model was trained to generate 16 video frames at a resolution of 512x320
given a context frame of the same resolution.

• Developed by: CUHK & Tencent AI Lab
• Funded by: CUHK & Tencent AI Lab
• Model type: Video Diffusion Model
• Finetuned from model: DynamiCrafter-interpolation (512x320)

Model Sources

For research purpose, we recommend our Github repository (https://github.com/ToonCrafter/ToonCrafter),
which includes detailed implementations.

• Repository: https://github.com/ToonCrafter/ToonCrafter
• Paper: https://arxiv.org/abs/2405.17933
• Project page: https://doubiiu.github.io/projects/ToonCrafter/
• Demo1: https://huggingface.co/spaces/Doubiiu/tooncrafter
• Demo2: https://replicate.com/fofr/tooncrafter

Uses

Feel free to use it under the Apache-2.0 license. Note that we don't have any official commercial product for ToonCrafter currently.

Limitations

• The generated videos are relatively short (2 seconds, FPS=8).
• The model cannot render legible text.
• The autoencoding part of the model is lossy, resulting in slight flickering artifacts.

How to Get Started with the Model

Check out https://github.com/ToonCrafter/ToonCrafter