1 年 以前
(E)
【stable diffusion】step2: AI绘图安装部署,零基础入门
以上两种方法任选一种:
方法1:
A1111 Webui Launcher 下载地址
https://kgithub.com/Empire...
方法2:(推荐)
秋叶一键启动V2(含Stable Diffusion)下载地址
夸克链接(新):https://pan.quark.cn/s/790...
百度链接(旧):https://pan.baidu.com/s/12...
提取码:dhua
三个压缩包下载后放一起解压即可。
基础模型存放地址:
novelai-webui-aki-v2modelsStable-diffusion
novelai-webui-aki-v2modelsStable-diffusion
#sd模型
以上两种方法任选一种:
方法1:
A1111 Webui Launcher 下载地址
https://kgithub.com/Empire...
方法2:(推荐)
秋叶一键启动V2(含Stable Diffusion)下载地址
夸克链接(新):https://pan.quark.cn/s/790...
百度链接(旧):https://pan.baidu.com/s/12...
提取码:dhua
三个压缩包下载后放一起解压即可。
基础模型存放地址:
novelai-webui-aki-v2modelsStable-diffusion
novelai-webui-aki-v2modelsStable-diffusion
#sd模型
1 年 以前
最新入驻 | FilmGirl 胶片风 Film Grain LoRA
模型原创者 LEOSAM
Stable Diffusion lora模型下载
https://www.liblibai.com/#...
#sd模型
#lora模型
模型原创者 LEOSAM
Stable Diffusion lora模型下载
https://www.liblibai.com/#...
#sd模型
#lora模型
1 年 以前
1 年 以前
【AI辅助建模】将StableDiffusion生成的2D设计图片
01:Point-E 算法
https://huggingface.co/spa...
02:ZoeDepth 深度图转3D模型
https://huggingface.co/spa...
03:Monster Mash
https://monstermash.zone/
01:Point-E 算法
https://huggingface.co/spa...
02:ZoeDepth 深度图转3D模型
https://huggingface.co/spa...
03:Monster Mash
https://monstermash.zone/
Hugging Face – The AI community building the future.
We’re on a journey to advance and democratize artificial intelligence through open source and open science.
https://huggingface.co/spa...
1 年 以前
ai真人模型 练出自己的ai老婆
git仓库:https://github.com/AUTOMAT...
checkpoint文件拷贝到路径:stable-diffusion-webuimodelsStable-diffusion
lora文件拷贝到路径:stable-diffusion-webuimodelsLora
texture文件拷贝到路径:stable-diffusion-webuiembeddings
git仓库:https://github.com/AUTOMAT...
checkpoint文件拷贝到路径:stable-diffusion-webuimodelsStable-diffusion
lora文件拷贝到路径:stable-diffusion-webuimodelsLora
texture文件拷贝到路径:stable-diffusion-webuiembeddings
GitHub - AUTOMATIC1111/stable-diffusion-webui: Stable Diffusion web UI
Stable Diffusion web UI. Contribute to AUTOMATIC1111/stable-diffusion-webui development by creating an account on GitHub.
https://github.com/AUTOMATIC1111/stable-diffusion-webui.git
1 年 以前
安裝Stable Diffusion下載軟件鏈接 :
【Windows版GIT】https://gitforwindows.org
【Python 3.10.8】https://www.python.org/dow...
【AUTOMATIC1111】https://github.com/AUTOMAT...
【Stable Diffusion 1.5模型】https://huggingface.co/run...
【Windows版GIT】https://gitforwindows.org
【Python 3.10.8】https://www.python.org/dow...
【AUTOMATIC1111】https://github.com/AUTOMAT...
【Stable Diffusion 1.5模型】https://huggingface.co/run...
1 年 以前
(E)
全球最快的开源绘画模型
OneFlow Stable Diffusion 使用地址:https://github.com/Oneflow...
OneFlow 地址:https://github.com/Oneflow...
#ai模型
OneFlow Stable Diffusion 使用地址:https://github.com/Oneflow...
OneFlow 地址:https://github.com/Oneflow...
#ai模型
How to Run OneFlow Stable Diffusion · Oneflow-Inc/diffusers Wiki · GitHub
OneFlow backend for 🤗 Diffusers. Contribute to Oneflow-Inc/diffusers development by creating an account on GitHub.
https://github.com/Oneflow-Inc/diffusers/wiki/How-to-Run-OneFlow-Stable-Diffusion
1 年 以前
ai紫色眼镜写实美女提示词实例
woman with flowers on her head,sketch,Caustics,hime cut,bewitching,landscape,deep,in the rain,beautiful purple sunset at beach in the ocean,rose petals,in the baroque architecture ,raw photo, photorealistic, masterpiece, best quality, hires, intricate details, reflections, photographed on a Canon EOS R5, 50mm lens, F/2.8, HDR, 8k resolution, ray-tracing
#ai绘画提示词
#Stable
woman with flowers on her head,sketch,Caustics,hime cut,bewitching,landscape,deep,in the rain,beautiful purple sunset at beach in the ocean,rose petals,in the baroque architecture ,raw photo, photorealistic, masterpiece, best quality, hires, intricate details, reflections, photographed on a Canon EOS R5, 50mm lens, F/2.8, HDR, 8k resolution, ray-tracing
#ai绘画提示词
#Stable
1 年 以前
ai动漫女孩提示词实例
A charming chibi girl with anime-style eyes poses for a portrait, her intricate hair beautifully rendered. A subtle shimmer in the air adds a touch of magic to the symmetrical composition. The concept art features digital painting with crisp lines to capture the essence of the re:Zero style. The girl looks directly into the camera, making the viewer feel engaged. This square image is perfect for a social media profile. For this stunning portrait, the best camera would be a high-resolution DSLR, with a prime lens that can capture intricate details with a shallow depth of field. The lighting should be bright and even. Using Adobe Photoshop would be an excellent way to create a distinctive effect, like adding a subtle glow. Recommended equipment: Canon EOS 5D Mark IV, 50 mm f/1.4 USM lens, LED lights, Adobe Photoshop.
#ai绘画提示词
#Stable
A charming chibi girl with anime-style eyes poses for a portrait, her intricate hair beautifully rendered. A subtle shimmer in the air adds a touch of magic to the symmetrical composition. The concept art features digital painting with crisp lines to capture the essence of the re:Zero style. The girl looks directly into the camera, making the viewer feel engaged. This square image is perfect for a social media profile. For this stunning portrait, the best camera would be a high-resolution DSLR, with a prime lens that can capture intricate details with a shallow depth of field. The lighting should be bright and even. Using Adobe Photoshop would be an excellent way to create a distinctive effect, like adding a subtle glow. Recommended equipment: Canon EOS 5D Mark IV, 50 mm f/1.4 USM lens, LED lights, Adobe Photoshop.
#ai绘画提示词
#Stable
1 年 以前
ai动漫人物提示词
Model: Stable Diffusion
Scale: 7
Steps: 25
Width: 768
Height: 512
" anime girl, blue eyes, portrait in extremely cute realistic digital art style with bloom ethereal effects, trending on Artstation, pixiv, made by WLOP, feminine in pastel shades, hyperdetailed, 4k, 8k, ultra HD "
#ai绘画提示词
Model: Stable Diffusion
Scale: 7
Steps: 25
Width: 768
Height: 512
" anime girl, blue eyes, portrait in extremely cute realistic digital art style with bloom ethereal effects, trending on Artstation, pixiv, made by WLOP, feminine in pastel shades, hyperdetailed, 4k, 8k, ultra HD "
#ai绘画提示词
1 年 以前
ai动漫人物提示词
Model: Stable Diffusion (v1.5)
Scale: 7
Steps: 25
Sampler: DPM Solver++
Width: 768
Height: 512
" ******blurred oil painting,the lilith looks bewitching ,a portrait of beautiful kawaii anime girl from AzurLane, by kaede, by krenz cushart, A girl against the backlight at dusk,knee shot,detailed face, eyes twincle,hyper angle pose, dynamic angle,wide angle,mini dress and mini skirt ,trending of pixiv fanbox,Violet Evergarden of Kyoto Animation,makoto shinkai, takashi takeuchi, Studio Ghibli, 8k,HQ "
#ai绘画提示词
Model: Stable Diffusion (v1.5)
Scale: 7
Steps: 25
Sampler: DPM Solver++
Width: 768
Height: 512
" ******blurred oil painting,the lilith looks bewitching ,a portrait of beautiful kawaii anime girl from AzurLane, by kaede, by krenz cushart, A girl against the backlight at dusk,knee shot,detailed face, eyes twincle,hyper angle pose, dynamic angle,wide angle,mini dress and mini skirt ,trending of pixiv fanbox,Violet Evergarden of Kyoto Animation,makoto shinkai, takashi takeuchi, Studio Ghibli, 8k,HQ "
#ai绘画提示词
1 年 以前
ai动漫人物提示词
Model: Stable Diffusion (v1.5)
Scale: 7
Steps: 25
Sampler: DPM Solver++
Width: 768
Height: 512
"beautiful kawaii girl, kawaii, large blue eyes, wavy Medium-long length silver hair, dressed in kimono, exposed shoulders, focus on face, full body, glamorous body, legs, high angle, cygames, shadowverse, painted by mushimaro, game character, line art, thickly painted, hyper angle pose, trending on pixiv, detailed, illustration, natural lighting, recent photograph or photography "
#ai绘画提示词
Model: Stable Diffusion (v1.5)
Scale: 7
Steps: 25
Sampler: DPM Solver++
Width: 768
Height: 512
"beautiful kawaii girl, kawaii, large blue eyes, wavy Medium-long length silver hair, dressed in kimono, exposed shoulders, focus on face, full body, glamorous body, legs, high angle, cygames, shadowverse, painted by mushimaro, game character, line art, thickly painted, hyper angle pose, trending on pixiv, detailed, illustration, natural lighting, recent photograph or photography "
#ai绘画提示词
1 年 以前
ai动漫人物提示词
Model: Stable Diffusion (v1.5)
Scale: 7
Steps: 25
Sampler: DPM Solver++
(寬)Width: 768
(高)Height: 512
提示词: "beautiful girl, moe style, large deep clear eyes, straight long length white hair, medieval dress, exposed shoulders, focus on face, full body, glamorous body, legs, high angle, hyper angle pose, trending on pixiv, detailed, illustration, shadowverse, granblue fantasy, cygames, mushimaro, anime detailed line work, game character, comic cover, volumetric lighting, atmospheric lighting "
#ai绘画提示词
Model: Stable Diffusion (v1.5)
Scale: 7
Steps: 25
Sampler: DPM Solver++
(寬)Width: 768
(高)Height: 512
提示词: "beautiful girl, moe style, large deep clear eyes, straight long length white hair, medieval dress, exposed shoulders, focus on face, full body, glamorous body, legs, high angle, hyper angle pose, trending on pixiv, detailed, illustration, shadowverse, granblue fantasy, cygames, mushimaro, anime detailed line work, game character, comic cover, volumetric lighting, atmospheric lighting "
#ai绘画提示词
1 年 以前
ai绘画提示词的一些使用技巧
一、画质区:用来提升整个图片的画质,图片的质量与风格。8k, best quality(8K,最佳品质), masterpiece(杰作),ultra highres(超高分辨率),RAW photo,(RAW照片)等等这些提示词用到10个以内可以明显提升画出的质感。
高质量画质
二、人物的写实的画质描述:1 girl,(1个女孩)detailed beautiful skin,(细致美丽的肌肤)kind smile,(亲切的微笑)absurdres,(夸张)detailed-beautiful-face,(精致美丽的容颜)petite figure,(身材娇小)detailed skin texture,(细致的皮肤纹理)pale skin,(白皙肌肤)detailed hair,(细致的头发) random hair style,(随意发型,)detailed eyes,(细致的眼睛)glistening skin,(晶莹剔透肌肤)。clean face,pretty face,sexy, cool, naughty face,((twintails)),这些我就不翻译大家可以自行百度。其中twintails用了两个括弧是权重加强。正常的情况下必出两个辫子。
用双括弧来加强权重((twintails))
三、角色描述区:衣服、发型、肤色、表情都可以放在这个区。short sleeve shirts,(短袖衬衫)uniform,(制服)pantyhose,(连裤袜)pleated skirt,(百褶裙)ear rings,(耳环)futuristic,(未来派)studio,(工作室)white background,(白色背景)这些都可以进行细划。耳环你可以更具体的描术。这样Ai出的图会更接你想要的效果。
四、用L ora的模型来控制人物的脸型。如果只想用到基础的模型的话也可以不用。要用到Lora的模型时总的权重不能超过1.超过来就会出来奇怪的画出。.0.4+0.1+0.1=0.6三个Lora模型的权重相加没超过1是没有问题的。
#ai绘画提示词
#Stable
一、画质区:用来提升整个图片的画质,图片的质量与风格。8k, best quality(8K,最佳品质), masterpiece(杰作),ultra highres(超高分辨率),RAW photo,(RAW照片)等等这些提示词用到10个以内可以明显提升画出的质感。
高质量画质
二、人物的写实的画质描述:1 girl,(1个女孩)detailed beautiful skin,(细致美丽的肌肤)kind smile,(亲切的微笑)absurdres,(夸张)detailed-beautiful-face,(精致美丽的容颜)petite figure,(身材娇小)detailed skin texture,(细致的皮肤纹理)pale skin,(白皙肌肤)detailed hair,(细致的头发) random hair style,(随意发型,)detailed eyes,(细致的眼睛)glistening skin,(晶莹剔透肌肤)。clean face,pretty face,sexy, cool, naughty face,((twintails)),这些我就不翻译大家可以自行百度。其中twintails用了两个括弧是权重加强。正常的情况下必出两个辫子。
用双括弧来加强权重((twintails))
三、角色描述区:衣服、发型、肤色、表情都可以放在这个区。short sleeve shirts,(短袖衬衫)uniform,(制服)pantyhose,(连裤袜)pleated skirt,(百褶裙)ear rings,(耳环)futuristic,(未来派)studio,(工作室)white background,(白色背景)这些都可以进行细划。耳环你可以更具体的描术。这样Ai出的图会更接你想要的效果。
四、用L ora的模型来控制人物的脸型。如果只想用到基础的模型的话也可以不用。要用到Lora的模型时总的权重不能超过1.超过来就会出来奇怪的画出。.0.4+0.1+0.1=0.6三个Lora模型的权重相加没超过1是没有问题的。
#ai绘画提示词
#Stable
1 年 以前
ai欧美性感美女描述词
Art by Greg Rutkowski Instagram model ("Hannah Palmer"), blonde hair, blue eyes, large eyes, simple white bikini, rooftop background, intricate detail of skin, Danielle Ayala, Abigail Ratchford, Alexa Pearl, Demi Rose, fantasy of waitress in tavern
#ai绘画提示词
#midjourney
#Stable Diffusion
Art by Greg Rutkowski Instagram model ("Hannah Palmer"), blonde hair, blue eyes, large eyes, simple white bikini, rooftop background, intricate detail of skin, Danielle Ayala, Abigail Ratchford, Alexa Pearl, Demi Rose, fantasy of waitress in tavern
#ai绘画提示词
#midjourney
#Stable Diffusion
2 年 以前
最近可以在各个平台上看到stable diffusion的相关案例,各种text2img或是img2img的应用都有非常惊艳的效果。
虽然类似功能的DALL-E已经发布很久了,甚至性能更为强大,但是stable diffusion模型开源所带来的影响力是DALL-E所远远不能比拟的,投石入河激起的涟漪层层荡开,许多新的服务、网站衍生而出,不禁让人吃惊:怎么会那么火?
#ai绘画
https://zhuanlan.zhihu.com...
虽然类似功能的DALL-E已经发布很久了,甚至性能更为强大,但是stable diffusion模型开源所带来的影响力是DALL-E所远远不能比拟的,投石入河激起的涟漪层层荡开,许多新的服务、网站衍生而出,不禁让人吃惊:怎么会那么火?
#ai绘画
https://zhuanlan.zhihu.com...
stable diffusion安装与简单评测 - 知乎
最近可以在各个平台上看到stable diffusion的相关案例,各种text2img或是img2img的应用都有非常惊艳的效果。 虽然类似功能的DALL-E已经发布很久了,甚至性能更为强大,但是stable diffusion模型开源所带来的影响力…
https://zhuanlan.zhihu.com/p/563731965
主办单位
yua aida
1 年 以前