tifa
最近到了樱花季了，想到之前最火的一期是TIFA的图，
突然有了这个"TIFA × 樱花"专题的想法。
试了一下，最有意境的组合是
- 底模：dalcefoPainting v3
- LoRA：koreanDollLikeness + tifaMeenow_tifa
- 核心关键词：cherry blossom, city, Tokyo,((city pop)), idol, 1girl
其他的自由发挥，大家看成片吧，我会放一组参数在最后，直接一键复制到stable diffusion webui里
主模: dalcefoPainting_3rd
使用Lora模型Tifa：
参数（可直接复制到stable diffusion一键自动填充）
<(masterpiece, realistic:1.3), (extremely intricate:1.2)>, full body, nature, forest, trees, cherry blossom,
(tifa lockhart,red_eyes,black hair,long hair,) ,small breasts, white dress,
<lora:koreanDollLikeness_v10:0.4> , <lora:tifaMeenow_tifaV2:0.9>
Negative prompt: (worst quality, low quality:1.4), (depth of field, blurry:1.2), (greyscale, monochrome:1.1), 3D face, cropped, lowres, text, jpeg artifacts, signature, watermark, username, blurry, artist name, trademark, watermark, title, multiple view, Reference sheet, curvy, plump, fat, muscular female, strabismus,
Steps: 26, Sampler: DPM++ 2M Karras, CFG scale: 10, Seed: 499369207, Size: 520x680, Model hash: 5efb10a9e1, Model: dalcefoPainting_3rd, Denoising strength: 0.5, Hires upscale: 1.5, Hires upscaler: Latent (nearest-exact)
示例
相关文章
Thank you very much for sharing. Your article was very helpful for me to build a paper on gate.io. After reading your article, I think the idea is very good and the creative techniques are also very innovative. However, I have some different opinions, and I will continue to follow your reply.
The point of view of your article has taught me a lot, and I already know how to improve the paper on gate.oi, thank you. https://www.gate.io/th/signup/XwNAU
The point of view of your article has taught me a lot, and I already know how to improve the paper on gate.oi, thank you. https://www.gate.io/ru/signup/XwNAU
Reading your article has greatly helped me, and I agree with you. But I still have some questions. Can you help me? I will pay attention to your answer. thank you.