Toonify: Turn Real Faces into Animated Disney Characters

Key Takeaways

this ai can transform any of your pictures into accurate representation with a disney animated movie character style you can even use it on yourself right now with a simple click using tunify it's the first link in the description but before testing it out let's quickly see how it works [Music] this is what's ai and i share artificial intelligence news every week if you are new to the channel and want to stay up to date please consider subscribing to not miss any further news using a publicly available pre-trained model called style gun 2 they were able to create tunify a website transforming any picture taken from profile into a disney pixar or dreamworks animated movie character style gun 2 is a can architecture model made by the nvidia research team it currently has state of the art results in generative image modeling in short they use the original stargane architecture and improved it to improve the style related results this network is a generative adversarial network merged with style transfer style transfer is a technique used to change the style of a whole image based on the different styles it was trained on as you can see here the generative adversarial network what we call gun is then the base architecture of the network it's basically composed of two main parts at first there's what we call a generator network which learns high level attributes of the human faces it is trained on such as pose and identity of the person to control the image synthesis process this is where it learns the style of the images fed such as a human style or a disney cartoon style then to train this generator we need the second part of the gan architecture which is a discriminator in short both generated and real images are fed to the discriminator during training which tries to discriminate the real images from the generated ones based on this comparison's feedback the generator network can iteratively improve the quality of the fake images and fool the discriminator as i said to nfi used a pre-trained model which is a model that has already been trained on many images yielding good results then they fine-tuned it by training their generator again with 300 more animated movie character pictures related to disney pixar and dreamworks world then merging specific parts of two different game networks such as one trained on human faces and another one trained on animated movie characters they were able to create a hybrid which has the structure of a cartoon face but photorealistic rendering this gives a high quality cartoon style look to any person and not any cartoon style but the one it has been trained on which in this case are disney pixar and dreamworks movie characters you could do these same steps on any other cartoon styles if you have enough images of this specific style the one impressive thing here is what they've done by merging the two networks in short they understood that the first few layers were responsible for the pose of the head and shape of the face which is controlled by the human faces gun trained network then they saw that the latter layers were more about controlling things like lighting and texture which is what we are looking for with the cartoon images allowing them to generate images from an entirely novel domain and do this with a high degree of control meaning that they could conserve the property of the human face it was fed with and generate a cartoon version of it with a specific style such as disney characters rather than generating images with random properties that are not really following the shapes and the soul of the image fed like most approaches they call this the layer swapping technique in their paper which i linked in the description as well as the tunify website where you can try to cartoonize yourself or any picture you want the tunify website is the first link in the description if you would like to try it out on yourself or any picture you want but here's a quick note from the authors before testing it yourself to produce great results the original image must be a fairly high quality such as around 512 per 512 pixels for the face then side faces and faces of under 128 by 128 pixels are unlikely to be found by the network have fun and please share your results in the comment section with a link to your images or tag me on twitter using at whatsaii i would love to see them please leave a like if you went this far in the video and since there are over 90 percent of you guys watching that are not subscribed yet consider subscribing to the channel to not miss any further news clearly explained if you would like to start or improve with machine learning i've linked all the best online courses in a repository in the description thank you for watching [Music] you