尘白禁区-肴-冬至| Yao-winner|スノウブレイク：禁域降臨 | SNOWBREAK: CONTAINMENT ZONE | 塵白禁域| Yao版本v1.0 (ID: 151775)

请阅读使用说明，请使用google翻译,Please read the instructions, please use google translate,

説明書を読んで、Google翻訳を使用してください ,사용설명서를 읽고 Google 번역을 사용하세요.

描述:

权重：0.6-0.8，推荐0.7，权重越高，3D感越强

推荐使用：高分辨率修复 (Hires. fix)、controlnet-tile图生图重绘

【强烈推荐使用模型】：OnlyAnime |《唯》· 炫彩动漫 https://civitai.com/models/105955/onlyanime-or

【负面embaddings】: UnrealisticDream, EasyNegative

vae: kl-f8-anime2

——————————————————————————————————————————————————————

【Trigger Words】

1.正面提示词：yaoDZCB,1girl, black hair, long hair, ponytail, bodysuit, helmet,

正面可选提示词：(red_pink_eyes:0.7), thighhighs, high heels, single_bare_shoulder, pelvic curtain, gloves

2.负面提示词：3D,3D rendering

负面可选提示词：weapon, holding, holding weapon, rifle, gun, holster, poor light

3.负面embaddings: UnrealisticDream, EasyNegative,

【一个参考预设，您可以在此之上进行修改】

1.正面提示词：yaoDZCB,1girl, solo, black hair, long hair, ponytail, sparkling eyes, from behind, looking back, breasts, standing, bodysuit, helmet, thighhighs, single_bare_shoulder, pelvic curtain, cowboy shot, naughty face, (red_pink_eyes:0.7), masterpiece, best quality, (colorful),(finely detailed beautiful eyes and detailed face),cinematic lighting, extremely detailed CG unity 8k wallpaper, <lora:yaoDZCB-v1:0.7>

2.负面提示词：EasyNegative, UnrealisticDream,3D, 3D rendering, weapon, holding weapon, rifle, gun, holster, dark scene, poor light

3.其它参数：DPM++SED Karras, 40 steps ,512x1024， 启用Hires.fix, R-ESRGAN 4X+Anime6B, 放大1.2倍，重绘幅度0.35

——————————————————————————————————————————————————————

【目前存在的问题与建议】

1.（1）人物的眼睛与面部精度有概率出现瑕疵，但可以用以下方式【3】完美修复。

（2）bodysuit，pelvic curtain，single_bare_shoulder，gloves 有概率出现错误。

2.在绘制远景或全身画面时，请适当增大出图尺寸与步数。

3.如果以上方法【2】的效果不理想，请直接将生成的图片发送至图生图进行重绘【效果很好】，方法如下：

（1）尽量使用与原图一致的提示词、角色lora、seed

（2）启用controlnet--tile, 启用 放大脚本 Ultimate SD upscale

（3）一个参考设置参数：Steps:30-40, 重绘幅度0.3-0.4，放大算法：4x-UltraSharp 或 R-ESGAN 4x+anime6B, 放大倍数1.25--

4.在创作NSFW内容时，请根据实际情况降低权重至0.5-0.6，同时使用更多的NSFW相关提示词，增加NSFW相关提示词的权重，并将bodysuit,helmet等衣服类Trigger Words 加入负面提示词。

5.当您感觉图片细节不足、模糊时，请适当增加迭代步数，或调整采样器及vae,也可以尝试【3】的方法进行图生图重绘。

6.该Lora的权重越低，越应该提高人物特征类【Trigger Words】的权重，例如：yellow eyes, brown hair, ponytail

7.强烈推荐使用负面embaddings:UnrealisticDream, EasyNegative

8.可能存在的模型泛用性问题

——————————————————————————————————————————————————————————————————————————————————————————————————————————————————

Weight: 0.6-0.8, 0.7 is recommended, the weight, the stronger the 3D feeling

Recommended uses: high-resolution repair (Hires.fix), controlnet-tile image redrawing

[Strongly recommended to use models]: OnlyAnime | "only" · Colorful https://civitai.com/models/105955/onlyanime-or

[Negative Embedding]: UnrealisticDream, EasyNegative

_____________________________________________________________________________________________________________________________________

【Trigger Words】

1. Positive prompt words: yaoDZCB,1girl, black hair, long hair, ponytail, bodysuit, helmet,

Optional prompt words on the front: (red_pink_eyes:0.7), thighhighs, high heels, single_bare_shoulder, pelvic curtain, gloves

2. Negative prompt words: 3D, 3D rendering

Negative optional prompt words: weapon, holding, holding weapon, rifle, gun, holster, poor light

3. Negative embaddings: UnrealisticDream, EasyNegative,

[A reference preset that you can modify on top of]

1. Positive prompt words: yaoDZCB,1girl, solo, black hair, long hair, ponytail, sparkling eyes, from behind, looking back, breasts, standing, bodysuit, helmet, thighhighs, single_bare_shoulder, pelvic curtain, cowboy shot, naughty face , (red_pink_eyes:0.7), masterpiece, best quality, (colorful),(finely detailed beautiful eyes and detailed face),cinematic lighting, extremely detailed CG unity 8k wallpaper, <lora:yaoDZCB-v1:0.7>

2. Negative prompt words: EasyNegative, UnrealisticDream, 3D, 3D rendering, weapon, holding weapon, rifle, gun, holster, dark scene, poor light

3. Other parameters: DPM++SED Karras, 40 steps, 512x1024, enable Hires.fix, R-ESRGAN 4X+Anime6B, zoom in 1.2 times, redraw range 0.35

_____________________________________________________________________________________________________________________________________

[Current problems and suggestions]

1. (1) The accuracy of the eyes and face of the character may be flawed, but it can be perfectly repaired by the following method [3].

(2) bodysuit, pelvic curtain, single_bare_shoulder, gloves have a probability of error.

2. When drawing a distant view or a full-body picture, please increase the size and steps of the drawing appropriately.

3. If the effect of the above method [2] is not satisfactory, please directly send the generated picture to Tushengtu for redrawing [the effect is very good], the method is as follows:

(1) Try to use the prompt words, character lora, and seed that are consistent with the original picture

(2) Enable controlnet--tile, enable the zoom script Ultimate SD upscale

(3) A reference setting parameter: Steps:30-40, redrawing range 0.3-0.4, magnification algorithm: 4x-UltraSharp or R-ESGAN 4x+anime6B, magnification factor 1.25--

4. When creating NSFW content, please reduce the weight to 0.5-0.6 according to the actual situation, use more NSFW-related prompt words, increase the weight of NSFW-related prompt words, and add negative prompts to Trigger Words for clothing such as bodysuit, helmet, etc. word.

5. When you feel that the image details are insufficient or blurry, please increase the number of iteration steps appropriately, or adjust the sampler and VAE. You can also try the method [3] to redraw the image.

6. The lower the weight of the Lora, the more the weight of the character feature class [Trigger Words] should be increased, such as: yellow eyes, brown hair, ponytail

7. It is strongly recommended to use negative emaddings: UnrealisticDream, EasyNegative

8. Possible model generality issues

训练词语: yaoDZCB,1girl, black hair, long hair, ponytail, bodysuit, helmet, (red_pink_eyes:0.7), thighhighs, high heels, single_bare_shoulder, pelvic curtain, gloves