【LORA】FilmGirl 胶片风 Film Grain LoRA & LoHAFilmVelvia2[803F3FA8D7]

名称：FilmVelvia2.safetensors
类型：【LORA】
程序版本：SD 1.5
文件大小：144.11MB
概述：
This is LoHA & LoRA models for generating film style AI portrait photos. when used with majicMIX realistic、 MoonMIX or Chilloumix, it can generate realistic film style photos. At the same time, it is recommended to use the TEXTUAL INVERSION model: bad_pictures, negative_hand Negative Embedding. These two Negative Embedding models will not significantly affect the film texture.
这是用于生成胶片风格AI写真照片的LoHA & LoRA模型。与majicMIX realistic、MoonMIX或Chilloumix搭配使用，可以生成逼真的胶片风格照片。同时推荐搭配TEXTUAL INVERSION模型：bad_pictures、negative_hand Negative Embedding。这两个Negative Embedding模型不会显著影响到胶片的质感。
Velvia 2.0的主体LoRA模型的训练集图像分辨率进一步提升至768*1152。
本版本主体模型训练时，加入了15张亚洲年轻男性胶片训练照片，以提高模型在生成男性照片时的真实感。
在高分辨率主体模型基础上，少比例融合了之前两版评价良好的Filmgirl历史版本模型。经过多次测试，这样融合模型的操作可以减少LoRA模型对脸型的影响。
请注意：
我常用的CFG范围是6~7，Lora权重通常在0.5~0.7。
Clip skip设为1或2时，生成的效果会有所不同，大家可以多尝试看自己喜欢哪种风格。

Hash	parameter
SHA256	803F3FA8D735FE885F8D6D0CB79F6FC1A033B18AB05E0DD020C050D7423EBDA2
CRC32	82664B4B
BLAKE3	E1FA89FD82CD0C44BA9215AF8EF46D3D627F8F26FD2BCB9CDF756AB76FD971EE
AUTOV1	7FE2E2DC
AUTOV2	803F3FA8D7

作品展示：

Generation Data：

1. fashion photography portrait of woman avatar, 1girl in blue lush Alien Rainforest with flowers and birds, fantasy, octane render, hdr, Dolby Vision, (intricate details, hyperdetailed:1.2), (natural skin texture, hyperrealism, soft light:1.2), fluffy short hair, , sharp focus, night, necklace, Chinese mythology, cleavage, medium breasts, sci-fi headband, looking at viewer, best quality, perfect body, <lora:FilmVelvia2:0.6>
   
2. Negative prompt: cartoon, anime, sketches,(worst quality, low quality), (deformed, distorted, disfigured), (bad eyes, wrong lips, weird mouth, bad teeth, mutated hands and fingers:1.2), bad anatomy, wrong anatomy, amputation, extra limb, missing limb, floating limbs, disconnected limbs, mutation, ugly, disgusting, (bad_pictures, negative_hand-neg:1.2)
   
3. Steps: 27, Size: 512x768, Seed: 2127226176, Model: majicmixRealistic_v4, Version: v1.3.1, Sampler: DPM++ 2M Karras, CFG scale: 7, Clip skip: 2, Model hash: d819c8be6b, Hires steps: 21, "FilmVelvia2: 142628a09ce9", Hires upscale: 2, Hires upscaler: Latent (bicubic antialiased), Denoising strength: 0.58

复制代码

Generation Data：

1. <lora:FilmVelvia2:0.6>, a woman standing on the side of a road at night, 1girl, long hair, motor vehicle, car, looking at viewer, ground vehicle, night, hands in pockets, blurry background, coat, black hair, parted lips, bokeh, jacket, brown hair, outdoors, red lips, upper body, artist name
   
2. Negative prompt: cartoon, anime, sketches,(worst quality, low quality), (deformed, distorted, disfigured), (bad eyes, wrong lips, weird mouth, bad teeth, mutated hands and fingers:1.2), bad anatomy, wrong anatomy, amputation, extra limb, missing limb, floating limbs, disconnected limbs, mutation, ugly, disgusting, (bad_pictures, negative_hand-neg:1.2)
   
3. Steps: 27, Size: 512x768, Seed: 3107565070, Model: majicmixRealistic_v4, Version: v1.3.1, Sampler: DPM++ 2M Karras, CFG scale: 7, Clip skip: 2, Model hash: d819c8be6b, Hires steps: 21, "FilmVelvia2: 142628a09ce9", Hires upscale: 2, Hires upscaler: Latent (bicubic antialiased), Denoising strength: 0.5

复制代码

Generation Data：

1. <lora:FilmVelvia2:0.6>, 1girl, solo, outdoor, cute japanese model girl, kimono, floral print, hair ornament, looking at viewer, hair flower, brown eyes, bangs, masterpiece, best quality, realistic
   
2. Negative prompt: cartoon, anime, sketches,(worst quality, low quality), (deformed, distorted, disfigured), (bad eyes, wrong lips, weird mouth, bad teeth, mutated hands and fingers:1.2), bad anatomy, wrong anatomy, amputation, extra limb, missing limb, floating limbs, disconnected limbs, mutation, ugly, disgusting, (bad_pictures, negative_hand-neg:1.2)
   
3. Steps: 27, Size: 512x768, Seed: 847472245, Model: majicmixRealistic_v4, Version: v1.3.1, Sampler: DPM++ 2M Karras, CFG scale: 7, Clip skip: 2, Model hash: d819c8be6b, Hires steps: 21, "FilmVelvia2: 142628a09ce9", Hires upscale: 2, Hires upscaler: Latent (bicubic antialiased), Denoising strength: 0.5

复制代码

Generation Data：

1. masterpiece, best quality, 1girl, solo,
   
2. sexy pose, pensive woman, voluminous dress, intricate lace, embroidered gloves, feathered hat, curled hairdo, pale skin, minimal makeup, tender smile, dainty neckline, nostalgic atmosphere, still life,
   
3. <lora:FilmVelvia2:0.6>
   
4. Negative prompt: cartoon, anime, sketches,(worst quality, low quality), (deformed, distorted, disfigured, bad eyes), wrong nose, weird mouth, strange ears, bad anatomy, wrong anatomy, amputation, extra limb, missing limb, floating limbs, (bad teeth, mutated hands, wrong fingers:1.2), disconnected limbs, mutation, ugly, disgusting, (bad_pictures:1.2), (negative_hand-neg:1.4)
   
5. Steps: 27, Size: 512x768, Seed: 332515045, Model: majicmixRealistic_v4, Version: v1.3.1, Sampler: DPM++ 2M Karras, CFG scale: 7, Clip skip: 2, Model hash: d819c8be6b, Hires steps: 21, "FilmVelvia2: 142628a09ce9", Hires upscale: 2, Hires upscaler: Latent (bicubic antialiased), Denoising strength: 0.6

复制代码

Generation Data：

1. Poetic lens,obscure light and shadow,aestheticism,somber film,1girl,emotional depth,dress shirt,flower,flower field,roses,bloom,<lora:Asian girls faceV2:0.5:face> <lora:FilmVelvia2:0.3>
   
2. Negative prompt: (ng_deepnegative_v1_75t) (badhandv4)
   
3. Steps: 30, Size: 1024x1536, Seed: 2239036174, Model: majicmixRealistic_v5, Version: v1.3.1, Sampler: DPM++ 2M alt Karras, CFG scale: 7, Clip skip: 2, Mask blur: 4, Model hash: 33c9f6dfcb, FilmVelvia2: 142628a09ce9", Denoising strength: 0.01, "Asian girls faceV2: 1b5796b08629

复制代码

下载地址：

游客，如果您要查看本帖隐藏内容请回复

感谢分享