Screenshot: AIP RIKEN

Nijigenka (二次元化) is Japanese for “2D-ification.” It’s a term for when artists take photos of people and turn them into anime characters. But now researchers are starting to have A.I. do the image-to-image translation.

At a recent workshop about deep learning in Tokyo, a researcher from Fudan University gave a presentation on creating anime characters with artificial intelligence. Researchers are using deep learning to produce anime versions of real photos. (For more image-to-image translations, watch this clip.)

Screenshot: AIP RIKEN

This poses challenges for anime because there are certain features that are specific to characters.

Advertisement

Screenshot: AIP RIKEN

Some results haven’t been that successful, which is why the researchers had to tune their method.

Screenshot: AIP RIKEN

Advertisement

For greater success.

Screenshot: AIP RIKEN

Expect this tech to get better and better.

Screenshot: AIP RIKEN

Advertisement


Kotaku East is your slice of Asian internet culture, bringing you the latest talking points from Japan, Korea, China and beyond. Tune in every morning from 4am to 8am.