赞
踩
SDXL有千般好,但是大显存却绕不开。更何况很多优秀的模型目前只有SD1.5版本的,今天就来介绍一些不同领域内好用的代表性的SD1.5模型,足够新人探索一段时间喽。(相关模型去C站搜标题找或者直接文末扫马领取)
目录
好看到不真实的写实风:majicMIX mrealistic 麦橘写实
更加真实的世界 realistic
幻想的世界:DreamShaper
超精细的日漫二次元:Counterfeit-V3.0
非日漫二次元的动画风格 Lyriel
B端ICON风格图片:DDicon
干净的产品图:Product Design
室内设计,房屋装修渲染:XSarchitectural-InteriorDesign
优点:人物,风景唯美,简单的提示词也能出高质量的美图。
缺点:过于精致完美,虽说是写实风的,但更像是写实风的CG感。而且描绘人物很难摆脱亚洲审美女性样貌(或许对很多人不是缺点?)
Sampler: Euler a, Euler, restart
Steps: 20~40
Hires upscaler: ESRGAN 4x or 4x-UltraSharp or 8x_NMKD-Superscale_150000_G
Hires upscale: 2+
Hires steps: 15+
Hires denoising strength: 0.05~0.5
clip skip 2
注意:不要开脸部修复(Face Restoration)!
样图参考:
1girl,face,curly hair,red hair,white background,
Negative prompt: (worst quality:2),(low quality:2),(normal quality:2),lowres,watermark,
Steps: 20, Sampler: Restart, CFG scale: 7, Seed: 2287632957, Size: 512x768, Model hash: 7c819b6d13, Model: majicmix7, VAE hash: 2453b80bc1, VAE: vae-ft-mse-840000-ema-pruned.safetensors, Denoising strength: 0.1, Clip skip: 2, ADetailer model: face_yolov8n.pt, ADetailer confidence: 0.3, ADetailer dilate/erode: 4, ADetailer mask blur: 4, ADetailer denoising strength: 0.08, ADetailer inpaint only masked: True, ADetailer inpaint padding: 24, ADetailer use inpaint width/height: True, ADetailer inpaint width: 512, ADetailer inpaint height: 512, ADetailer use separate sampler: True, ADetailer sampler: Euler a, ADetailer version: 23.9.3, Hires upscale: 2, Hires steps: 20, Hires upscaler: ESRGAN_4x, BMAB noise alpha final: 0.05, BMAB contrast: 0.9, BMAB brightness: 1.1, BMAB color temperature: 15, Version: v1.6.0
PS:麦橘还有其他系列也不错哦
优点:人物肖像更加真实,不再是完美的CG人,在还原真实世界上的SD1.5模型中很少有比他做的好的。
缺点:想要出好看的图片更考验各位炼丹师的手法了,而且人物基本上非亚洲人样貌。
Euler A or DPM++ SDE Karras
CFG Scale 3,5 - 7
Hires. fix with 4x-UltraSharp upscaler
Denoising strength 0.25-0.45
Upscale by 1.1-2.0
Clip Skip 1-2
ENSD 31337
触发词:analog style、modelshoot style
Prompt:
RAW photo, subject, 8k uhd, dslr, soft lighting, high quality, film grain, Fujifilm XT3
Negative Prompt:
(deformed iris, deformed pupils, semi-realistic, cgi, 3d, render, sketch, cartoon, drawing, anime), text, cropped, out of frame, worst quality, low quality, jpeg artifacts, ugly, duplicate, morbid, mutilated, extra fingers, mutated hands, poorly drawn hands, poorly drawn face, mutation, deformed, blurry, dehydrated, bad anatomy, bad proportions, extra limbs, cloned face, disfigured, gross proportions, malformed limbs, missing arms, missing legs, extra arms, extra legs, fused fingers, too many fingers, long neck, UnrealisticDream
(deformed iris, deformed pupils, semi-realistic, cgi, 3d, render, sketch, cartoon, drawing, anime, mutated hands and fingers:1.4), (deformed, distorted, disfigured:1.3), poorly drawn, bad anatomy, wrong anatomy, extra limb, missing limb, floating limbs, disconnected limbs, mutation, mutated, ugly, disgusting, amputation, UnrealisticDream
优点:适合生成有真实基础的幻想风格,包括不限于西幻,朋克,科技等等的肖像。就像另一个世界一样。(有SDXL版本)
缺点:新手不好把握提示词
ENSD 31337
Hires denoising strength: 0.25~0.6
clip skip 2
不开脸部修复(Face Restoration)
优点:日漫二次元的风格拿捏的死死的,小味一下就上来了。
第一张参考图片:
(masterpiece, best quality), lens flare, 1boy, male focus, classroom, solo, desk, indoors, shirt, school desk, black hair, white shirt, chair, window, collared shirt, school uniform, standing, short hair, pants, long sleeves, black pantsNegative prompt: EasyNegativeV2Steps: 25, Sampler: DPM++ 2M Karras, CFG scale: 10, Seed: 1599380747, Size: 1024x512, Model hash: db6cd0a62d, Model: CF5_Counterfeit-V3.0_fix_fix_fix, Denoising strength: 0.6, Clip skip: 2, ENSD: 31337, Hires upscale: 2, Hires upscaler: Latent
优点:大尺寸动画人物肖像很不错,不需要太多的提示词就有不错的细节,其他类型图片的其实也能用。光影方面的效果尤其不错。
缺点:作者被BAN了
DPM++ 2M Karras,
Clip skip 2
Sampler, Steps: 25-35+
样图:
(dark shot:1.1), epic realistic, portrait of halo, sunglasses, blue eyes, tartan scarf, white hair by atey ghailan, by greg rutkowski, by greg tocchini, by james gilleard, by joe fenton, by kaethe butcher, gradient yellow, black, brown and magenta color scheme, grunge aesthetic!!! graffiti tag wall background, art by greg rutkowski and artgerm, soft cinematic light, adobe lightroom, photolab, hdr, intricate, highly detailed, (depth of field:1.4), faded, (neutral colors:1.2), (hdr:1.4), (muted colors:1.2), hyperdetailed, (artstation:1.4), cinematic, warm lights, dramatic light, (intricate details:1.1), complex background, (rutkowski:0.66), (teal and orange:0.4)Negative prompt: 3d, cartoon, lowres, bad anatomy, bad hands, text, error, missing fingers, extra digit, fewer digits, cropped, worst quality, low quality, normal quality, jpeg artifacts, signature, watermark, username, blurry, artist name, young, loli, elf, 3d, illustrationSteps: 30, Sampler: DPM++ 2M Karras, CFG scale: 9, Seed: 1261263585, Size: 512x768, Model hash: e96852ac4a, Model: Lyriel_v16, Denoising strength: 0.56, Clip skip: 2, Hires resize: 832x1280, Hires steps: 38, Hires upscaler: Latent
优点:特定人群需要
缺点:特定人群需要
V1 版本细节多,v2 版本更简洁 结合DDICON_lora效果更好
样图:
masterpiece, best quality, super verbose, clear glass texture, frosted glass, Transparent sense of technology, dof, reflections, refractions, micro-details,studio lighting, Sunlight, blender, octane rendering, C4D, 2.5D, isometic render, (blue|white), White technology sense background, sharpen edges, DDICON, <lora:ddiconLora_v20:0.7>, (blue:1)Negative prompt: lowres, bad anatomy, ((bad hands)), (worst quality:2), (low quality:2), (normal quality:2), paintings, sketches, lowres, bad anatomy, bad hands, text, error, missing fingers, shadowSteps: 20, Sampler: Euler a, CFG scale: 7, Seed: 3106333066, Size: 512x512, Model hash: 1243b42854, Model: ddicon_v20, Clip skip: 2, ENSD: 31337, Version: v1.2.1, ControlNet 0: "preprocessor: canny, model: control_v11p_sd15_canny [d14c016b], weight: 1, starting/ending: (0, 1), resize mode: Crop and Resize, pixel perfect: True, control mode: Balanced, preprocessor params: (512, 82, 139)", ControlNet 1: "preprocessor: depth_zoe, model: control_v11f1p_sd15_depth [cfd03158], weight: 1, starting/ending: (0, 1), resize mode: Crop and Resize, pixel perfect: True, control mode: Balanced, preprocessor params: (512, 64, 64)"
优点:特定人群需要
缺点:不会ps,大部分生成的图片只能用来玩
Positive prompt:
3D product render, futuristic armchair, finely detailed, purism, ue 5, a computer rendering, minimalism, octane render, 4k
Negative prompt:
EasyNegative, (worst quality:2), (low quality:2), (normal quality:2), lowres, ((monochrome)), ((grayscale)), cropped, text, jpeg artifacts, signature, watermark, username, sketch, cartoon, drawing, anime, duplicate, blurry, semi-realistic, out of frame, ugly, deformed
Steps: from 20-40 (For EulerA is enough 20, DPM++SDE Karras or DPM++2S a)
CFG scale: 6-9 (8 Ideal).
触发词:3d product render、product render
优点:特定人群需要,有配合的lora使用效果更佳
缺点:特定人群需要
样图参考:
(masterpiece),(high quality), best quality, real,(realistic), super detailed, (full detail),(4k),8k,interior,Living room,white
Negative prompt: (normal quality), (low quality), (worst quality), paintings, sketches,
Steps: 20, Sampler: Euler a, CFG scale: 7, Seed: 2114630674, Size: 704x512, Model hash: 631eea1a0e, Model: XSarchitecturalV11InteriorDesgin, Denoising strength: 0.3, ENSD: 31337, Hires upscale: 2, Hires upscaler: R-ESRGAN 4x+
本文的所有相关模型,lora,样图已经整理好了,看文末扫马领取。
AIGC技术的未来发展前景广阔,随着人工智能技术的不断发展,AIGC技术也将不断提高。未来,AIGC技术将在游戏和计算领域得到更广泛的应用,使游戏和计算系统具有更高效、更智能、更灵活的特性。同时,AIGC技术也将与人工智能技术紧密结合,在更多的领域得到广泛应用,对程序员来说影响至关重要。未来,AIGC技术将继续得到提高,同时也将与人工智能技术紧密结合,在更多的领域得到广泛应用。
感兴趣的小伙伴,赠送全套AIGC学习资料和安装工具,包含AI绘画、AI人工智能等前沿科技教程,模型插件,具体看下方。
一、AIGC所有方向的学习路线
AIGC所有方向的技术点做的整理,形成各个领域的知识点汇总,它的用处就在于,你可以按照下面的知识点去找对应的学习资源,保证自己学得较为全面。
二、AIGC必备工具
工具都帮大家整理好了,安装就可直接上手!
三、最新AIGC学习笔记
当我学到一定基础,有自己的理解能力的时候,会去阅读一些前辈整理的书籍或者手写的笔记资料,这些笔记详细记载了他们对一些技术点的理解,这些理解是比较独到,可以学到不一样的思路。
四、AIGC视频教程合集
观看全面零基础学习视频,看视频学习是最快捷也是最有效果的方式,跟着视频中老师的思路,从基础到深入,还是很容易入门的。
五、实战案例
纸上得来终觉浅,要学会跟着视频一起敲,要动手实操,才能将自己的所学运用到实际当中去,这时候可以搞点实战案例来学习。
Copyright © 2003-2013 www.wpsshop.cn 版权所有,并保留所有权利。