基于Carl Zeiss Milvus镜头所拍摄照片训练的真实性风格。体现出蔡司镜头所拥有的景深、色彩、光斑、对比度等特性。适用于具有高度真实性的人物摄影照片。
触发词：Zeiss_style,
描述扩展：Depth of Field, Bokeh,
权重可以设置高一些。
Authentic style trained on photos taken with Carl Zeiss Milvus lenses. It reflects the depth of field, color,Bokeh, contrast and other characteristics of Zeiss lenses. Ideal for photographs of people with a high degree of realism.
Trigger words: Zeiss_style,
Description extensions: Depth of Field, Bokeh,
The weight can be set higher.
Description
FAQ
Comments (10)
一只猫为什么pureeroface啊
测试时候顺便输入的咒语，没认真看
not work for animal, all change to person if using this LORA. [Solved: remove pureeroface]
The pictures trained by the lora model do not contain people.Pay attention to write in the appropriate prompt. I made a mistake in the sample when I quickly tested lora. There should not be words that describe people, such as: skin, face, pores, freckles... You can fill in the animal alone to try.
@baldphotographer I just use your cat example, extra same model same seed same prompt same parameter. only different is I use A1111 default LORA but not addnet. It give me a girl with cat ear at the head. I delete all prompt relative to this LORA, a cat come out.
@z0_0d713 I accidentally added a "pureerosface_v1 "at the end of that example, which may cause characters to appear in different models, please delete it.
@baldphotographer Thanks! It work just remove pureeriface.
I like the idea of having a LoRA for each lense style, great work ❤️
Thank you so much. I'm going to try and make a lora with the AF 500 F8 Reflex lens from Minolta, which has an amazing donut-like lens bokeh.
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.