The neural network showed frames from the "Lord of the Rings" in the style of cartoons from Dreamworks

Reddit user Gagarin1961 used a neural network Stable Diffusionto turn footage from The Lord of the Rings movies into Pixar or Dreamworks cartoons. He shared his results with post, which collected 29 thousand likes in 5 days. The comments note that most of all this is similar to the work of Blue Sky Studios and How to Train Your Dragon from Dreamworks.

The author took frames from Peter Jackson's films as a basis, and also added a text description for generation in a given style. Here's what he got.

1. Gandalf

2. Frodo Baggins

3. Bilbo and Gandalf

4. Aragon

5. Gimli

6. Legolas

7. Bag End at Hobbiton

8. Saruman the White

9. Eye of Sauron

10. Walled city of Minas Tirith

11. "You shall not pass!"

12. Balrog

13. Rivendell

14. Galadriel

15. Elrond

In the comments, the author noted that not all the work was done by the neural network: in addition to selecting the proposed options, he also had to process some successful frames with minor flaws. He showed a couple of examples of raw images where something was wrong. Balrog initially scared in a completely different way:

And Gandalf initially did not stand with his back to the camera:

It's times like these that you really rejoice that Photoshop exists.