This LoRA generates a picture of Cardcaptor Sakura wearing her school uniform.
test prompt: 1girl, (lora:kinomotoSakuraSchool_20230311:0.6), kinomoto_sakura_school_uniforms
All training data are pictures drawn by AI. (but img2img is also included).
In the future, AI will learn from AI-drawn pictures.
I am making this future prediction.
And I made this LoRA to demonstrate it.
Perhaps this seems closer to embed(Textual Inversion).
It will allow you to use only a single keyword to display something that needs to be described with multiple prompt tags.
masterpiece, best quality, (cardcaptor_sakura) ,(kinomoto_sakura), brown hair, green eyes, (simple background), tomoeda elementary school uniform,school uniform , white skirt,miniskirt,white sailor collar, black jacket, black shirt,sailor collar,skirt, school uniform, hat, antenna hair, white headwear, white skirt, short hair, pleated skirt, serafuku, long sleeves, standing, neckerchief, sailor collar, black shirt, beret, blurry, shirt, white sailor collar, miniskirt, bangs
---> kinomoto_sakura_school_uniforms