Upload a photo and see what you'd look like in an animated movie!
Now in high definition.
ToonifyHD is currently a very early prototype which can generate images at much higher resolution than the original.
You can see some examples below, be sure to check them out at full 1024x1024 resolution!.
Reach out to me on Twitter or send an email to firstname.lastname@example.org
If you're looking for the original Toonify Yourself it's still available and free for everyone to use: Toonify Yourself.
This toonification system is made using deep learning. It uses a deep learning method called Generative Adversarial Networks. In fact it's a rather convoluted combination of StyleGAN2, resolution dependent model interpolation, and pixel2style2pixel.
At some point I'll write up more of the technical details on my blog.
This all started from some earlier experiments Doron shared on Twitter that got a lot of interest. For a description of those early experiments see this blog post on self toonification, the idea is the same, but the method is different (and not really suitable for hosting as a webapp).
We then made the original Toonify Yourself which got a lot of attention, and since then have been working to improve, producing Toonify HD.
Fundamentally this is all based on my work in Resolution dependent model interpolation, you can read the paper on Arxiv.
The algorithm works best with high resolution images without much noise. Looking straight on to the camera also seem to work best. Something like a corporate headshot tends to work well.
No your image is discarded as soon as you get your result. We don't store it or use it for anything other than producing your Toonification.
We use the open source dlib face detector to find faces, it's designed to pick up frontal faces but isn't perfect.
Not many characters from animated films wear glasses so the model seems to have learnt to mostly remove them. It also has problems with bald people, hats, and various other things.