This repo contains images of Kanji and radicals that users can generate in anime form with anime girls mostly, either though NovelAi and Waifu Diffusion.
The aim of this project is to bring WaniKani Mnemonics, primarily for meanings to life in Anime form with the power of AI.
To contribute, fork this repo, add images in the following format (kanji-(actualkanji).png/jpg/jpeg) and commit the repo. Then create a PR. Also, feel free to create a Greesemonkey script to make these images available for WaniKani reviews.
If you don't know how these images are made, watch this video about AI image generation by MKDHD
Follow these guidelines, or the merge PR will get rejected.
- Images must be safe for work
- Images should be close to the meaning mnemoic on WaniKani as much as possible as AI provides. You can add the reading meaning as well, but that is optional
- If animals are involved, use animal ears instead. AI for Anime image generation is bad at generating animals.
- Images should look acceptable and not break the immersion too much. Avoid images with extra limbs, too many obvious abnormalities, and disfigured faces and proportions.
- Female characters are preferred over male characters.
This guide in Japanese has keyboards how to customize how the resulting image will come out.
Since images generated by AI are not copyrightable and are not derivative works, any contributions are in the public domain so other learners of Japanese can use them as a study resource.