Stable Diffusion 实操教程：使用 LoRA 生成冰霜旗袍美女

三、提示词编写技巧

1. 正面关键词 (Positive Prompt)

包含画质权重、场景描述、人物特征及服装细节。

((masterpiece)),(highres:1.3), (((best quality))), ((ultra-detailed)), ((illustration)),(photorealistic:1.3),(realistic:1.4),(Cinematic Lighting:1.3),
dynamic angle, in the winter,(snow white forest:1.1), hazy mist,(on the hot springs:1.3) ,(many thin detailed snowflakes:1.3), (bloom), (shine), (twinkling stars:), (snowing), many white tree, detailed snowflakes,
cowboy shot,
(1 girl),(((solo))),running,(between legs:1.3),
(chinese girl:1.1),(big breasts:1.3),
hanfu,hanfugirl, beautiful detailed eyes,(black eyes:1.2),(pale firered hair:1.3),light smile,
,(beautiful detailed face:1.3),(cute), blushing, light smile, makeup ,lipstick,detailed clothes,(pale green transparent tight floral qipao dress:1.35),
(white fishnet thighhighs:1.25)

2. 负面关键词 (Negative Prompt)

用于排除常见缺陷和不想要的元素。

(cleavage:1.4),(Multiple people),(((simple background))),monochrome ,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, lowres, bad anatomy, bad hands, text, error, extra digit, fewer digits, cropped, worst quality, low quality, normal quality, jpeg artifacts, signature, watermark, username, blurry, ugly,pregnant,vore,duplicate,morbid,mut ilated,tran nsexual, hermaphrodite,long neck,mutated hands,poorly drawn hands,poorly drawn face,mutation,deformed,blurry,bad anatomy,bad proportions,malformed limbs,extra limbs,cloned face,disfigured,gross proportions, (((missing arms))),(((missing legs))), (((extra arms))),(((extra legs))),pubic hair, plump,bad legs,error,legs,username,blurry,bad feet ,(cleavage:1.4)

四、生成效果预览

通过上述配置，可生成具有冬季雪景氛围、穿着旗袍的人物图片。以下展示部分生成结果：

五、常见问题与优化建议

1. 人脸崩坏：尝试降低 CFG scale 至 5-6，或使用 FaceSwap 插件进行后期修复。
2. 细节模糊：开启 Hires Fix（高清修复），并适当增加 Steps 至 30 以上。
3. 颜色灰暗：检查 VAE 是否正确加载，或调整 Prompt 中的光照关键词（如 Cinematic Lighting）。
4. LoRA 失效：确认 LoRA 权重（Weight）设置是否合理，通常在 0.6-0.8 之间效果较佳。

通过不断调整上述参数与提示词，用户可以逐步掌握 Stable Diffusion 的生成逻辑，创作出符合预期的高质量图像。