[LoRA] Handdrawn

模型简介：
优化了整体线条，更加的流畅清晰，
非常好的泛化性，提示词控制画面即可
记得写触发词lineart，权重都可以尝试
低权重加强人物边缘轮廓线，高权重却越接近水彩笔画风
区别于线稿的地方是，线条更加粗一些，且画面精细度更高
不同的二次元底膜风格也不同，可以是水彩笔，纯漫画，钢笔等风格
例图中可以看到哦~
另外可以直接给画面增加颜色，权重降到0.7左右即可，或加强提示词权重
最后可以使用controlnet的lineart等进行人物形象绘制，例图中可见样版

Optimized the overall line, making it smoother and clearer,
Very good generalization, just control the screen with prompt words
Remember to write the trigger word 'lineart' and try weights
Low weight enhances character edge contours, while high weight approaches watercolor stroke style more closely
What sets it apart from the line draft is that the lines are thicker and the image is more precise
Different anime film styles are also different, which can be watercolor pen, pure cartoon, pen and other styles
As can be seen in the example picture~
In addition, you can directly add colors to the screen, reduce the weight to around 0.7, or enhance the weight of prompt words
Finally, you can use Controlnet's lineart and other tools to draw character images, as shown in the sample image

灵感来自一位某yin的主播，
简单的线条，略微偏向卡通的人物肖像。
单色，彩色均可，最开始目的是让大家快速绘制
类似水笔手绘风格的人物简易肖像，
尝试了各种比例，效果都还阔以，全身图建议开ADetailer
增加细节可以开启高清重绘或者其他lora，
人物肖像建议使用controlnet的lineart realisitc
权重一般建议0.7-1左右，也可尝试小权重，也许会有惊喜哦
所有例图均是封面图提示词直出，建议二次元底膜为主，
使用真人底膜出黑白图，例图中有实际效果，可以当个黑白滤镜

Inspired by a Tik Tok anchor,
Simple lines, slightly inclined towards cartoon character portraits.
Both monochrome and color options are available, with the initial goal of allowing everyone to quickly draw
A simple portrait of a character in a style similar to hand drawn water pen,
I have tried various proportions and the results are quite impressive. It is recommended to turn on ADetailer for the full body image
Adding details can enable high-definition redrawing or other Loras,
It is recommended to use Controlnet's lineart realism for character portraits
The recommended weight is generally around 0.7-1, or you can try small weights, which may bring surprises
All examples are directly displayed with the prompt words on the cover map. It is suggested that the anime bottom film be the main part,
Use a real person mask to create a black and white image. The example image has practical effects and can be used as a black and white filter
I found a new AI tool Shakker, a best image to image tool. You can try it via https://www.shakker.ai ,it can help you:
-Remix: Upload a picture. Just switch the prompts, and you can create stunning images in the same style.
-Style Transfer: Shakker not only extracts the style,but also switches among various styles.
Besides, Shakker also offers Object Control,Composition Control,Collage Redrawing etc.

~~~~~~~~~~~~~~~~~~~~~~~o(*￣︶￣*)o~~~~~~~~~~~~~~~~~~~~~~~

示例提示词和参数：
Prompt: Best quality,Masterpiece,1girl,Dragon,<lora:水彩笔v1.2:1>,lineart,

Negative prompt: (EasyNegative,ng_deepnegative_v1_75t,:1.2),(NSFW:2),(worst quality:2),(low quality:2),(normal quality:2),lowres,watermark,normal quality,,((grayscale)),skin spots,acnes,skin blemishes,age spot,(ugly:1.33),(duplicate:1.33),(morbid:1.21),(mutilated:1.21),(tranny:1.33),mutated hands,(poorly drawn hands:1.5),blurry,(bad anatomy:1.21),(bad proportions:1.33),messy hair,(hands:1.5),(glasses:1.5),Exposure,Chest,

Steps: 30, CFG scale: 15, Sampler: Euler a, Seed: 2025615733, Size: 512x768, Model: 二次元_天空之境 _ 苍玄NullStyle_V 2.0, Version: v1.4.0, CFG mode: Half Cosine Up, Model hash: c1d961233a, Hires steps: 15, Mimic scale: 7, Hires upscale: 1.5, Hires upscaler: 8x_NMKD-Superscale_150000_G, ADetailer model: face_yolov8s.pt, ADetailer version: 23.6.3, CFG scale minimum: 4, Denoising strength: 0.35, ADetailer mask blur: 4, ADetailer confidence: 0.3, Threshold percentile: 100, ADetailer dilate/erode: 4, "水彩笔v1.2: 8d5c161a81a9", ADetailer inpaint padding: 32, ADetailer denoising strength: 0.4, Dynamic thresholding enabled: True, ADetailer inpaint only masked: True