Models>

yingmo_璎陌——眷思量

masterpiece,best quality,,1girl,black hair, chinese clothes, <lora:yingmo:0.8>,outdoor,cowboy shot, dynamic pose, dynamic angle
masterpiece,best quality,,1girl,black hair, chinese clothes, <lora:yingmo:0.8>,outdoor,cowboy shot, dynamic pose, dynamic angle
masterpiece,best quality,,1girl,black hair, chinese clothes, <lora:yingmo:0.8>,outdoor,cowboy shot, dynamic pose, dynamic angle
Quick Generate
Select a style, write what you want to see and click Generate
What you want in the image.
Clear
Random
* This model is available for use and will generate similar character to the ones seen in the images.

Model Description by Creator

国漫眷思量中男主镜玄的母亲,拥有全剧最惊艳的建模和最少的镜头……从视频和官方海报搜集到的截图一共只有三十张,且均无高清图,视频建模精致,但是分辨率最高只支持1080P,打开显卡RTX视频增强的超分辨率,效果依然不是很理想,数据源的分辨率不高可能导致模型生成的细节不是很清晰,没有使用ai超分图训练是因为ai超分会导致原图失真,所以我的训练集一般都是用的原图,尽量还原原作风格,采用国风3 GuoFeng3 - v3.3 | Stable Diffusion Checkpoint | Civitai模型做为底模,基于秋叶的lora整合包进行训练权重在0.8左右效果比较稳定,配合EasyNegative等负面提示词模型效果更好。

权重在0.4时,能保留模型主体特征,并提供更大的自由度以更换姿势、服装及背景,画出更通用但不那么还原的图。

不开高清修复:我一般先使用文生图批量绘制低分辨率图,然后通过图生图SD upscale脚本进行超分辨率放大,540x960的图经过两次放大可得到效果更好的4k图片,此方法比直接高清修复更不吃显卡,可以使配置较低的硬件更快生成更高清的图片,放大时需要微调参数,因为是多张图放大后拼接,所以重绘幅度不宜过大,否则会导致图片不同部分毫无关联。动漫人物用真人底模绘制时,如果太还原可能没有那么好看,真人底模应该降低权重生成类似cosplay的图片。

开高清修复:初始分辨率540x960,采用R-ESRGAN 4x+放大算法,重绘幅度0.3放大2倍进行高清修复,然后图生图用SD upscale脚本,选择同样的算法,重绘幅度0.1进行超分放大,可以得到效果不错的4k高清图片。(此方法第一步直接生成1080P的图片,比较看脸,不易批量操作,16G显存最多只能同时生成4张图片,出图及修图效率较低,但是质量较高)

yingmo_璎陌——眷思量 - v1.0 | 吐司 [Blocked Link] (tusiart.com)(这里可以在线跑图看效果,需要开高清修复看脸抽卡,速度还可以,每天有免费额度)

Images Generated With This Model

Loading...

Train Models

Train AI Models On Your Own Pictures

You can use your own pictures to train AI models. Reimagine yourself in new ways with our AI-powered image generator.
Create pictures of yourself around the world, in your dream outfit, as a video game character, or in countless art styles. The possibilities are endless.
Train Your Own Model

Generate AI Images. Simple and Easy

AIEasyPic is the #1 AI image generator. Generate images with simple prompts, swap faces, train models to reimagine yourself, and more.