AI绘画——Stable Diffusion(待完成)

如今,AI绘画正在以飞一般的速度发展。笔者曾学过一小段时间的国画和板绘,会画一些简笔画。但是这些画的水平放在路人里还能说道说道,对比经过训练的画手那是相形见绌。而AI绘画的出现是一件很微妙的事情,将有专业门槛的技能下放给了普罗大众。这或许会让一些画师感到沮丧,但是面对变革,我们最好还是拥抱变化。

前置术语:
Prompt:
CFG scale:越高则遵循提示越多。

关键词一般设置在75个tokens。
常见关键词:https://stable-diffusion-art.com/how-to-come-up-with-good-prompts-for-ai-image-generation/#Some_good_keywords_for_you

Digital painting,Concept art
pixiv,sharp focus,expansive

颜色:muted color

custom models https://stable-diffusion-art.com/models/

为图像不同区域指定不同提示:
https://stable-diffusion-art.com/regional-prompter/

stable diffusion:https://huggingface.co/runwayml/stable-diffusion-v1-5
anything:https://huggingface.co/Linaqruf/anything-v3.0

安装模型:
https://stable-diffusion-art.com/models/#How_to_install_and_use_a_model

WebUI参数含义:
https://stable-diffusion-art.com/know-these-important-parameters-for-stunning-ai-images/#Seed

img2img:Prompt

two arms,cartoon,head without decoration,sleek helmet,muted colors,chibi style,smooth line

green,ugly, missing limbs,tiling, poorly drawn hands, poorly drawn feet, poorly drawn face, out of frame, disfigured, deformed, body out of frame, bad anatomy, watermark, signature, cut off, low contrast, underexposed, overexposed, bad art, beginner, amateur, distorted face, blurry, draft, grainy


如果有任何有错误或不够清晰的表达,欢迎邮件至 passacaglia@88.com,谢谢!

文章标题:AI绘画——Stable Diffusion(待完成)

字数:327

本文作者:Passacaglia

发布时间:2023-07-30, 23:44:51

最后更新:2023-09-23, 12:10:52

原始链接:https://passacaglia424.github.io/2023/07/30/AI%E7%BB%98%E7%94%BB/

版权声明: "署名-非商用-相同方式共享 4.0" 转载请保留原文链接及作者。