TJFilmStyle is a style LoRA I specifically trained to simulate real film photography, featuring the unique texture of physical film. It enhances the lighting, texture and close-up depth-of-field effects for landscapes, still lifes and portraits alike. Especially for portraits, it can significantly improve skin texture, replicate the translucent skin look of film photography, reduce the AI-generated artificial feel, and deliver more outstanding light and shadow performance.
Recommended Settings:
Trigger word: vifilm
This LoRA is trained on the Zib framework, and is compatible with both ZImage Base and ZImage Turbo series models.
Turbo model weight: 1.5-2; Base model weight: 0.6-0.8
Sampler: Euler, Scheduler: Simple
To get a more noticeable film grain effect, you can add the prompt word "film grain", select the "ddim_uniform" scheduler, or lower the denoise strength to around 0.88. It is not recommended to apply all three methods at the same time, to avoid image quality degradation.
It's not easy to train a style lora on zimage base. Sometimes V1 is not perfect on generating an obvious film effect, so that I'll continue training and releasing new versions, thanks for your support.
Prompt Tips: When generating images, try to avoid piling up and overlapping style-related prompt words, so as not to undermine the generation quality. It is recommended to use natural language descriptions instead.
TJFilmStyle是我专门训练，用来模拟真实胶片摄影的风格lora，带有胶片特有的质感，无论是风景、静物、还是人像，光影、质感以及近景的景深效果都得到了增强。尤其在人像方面，能明显提升皮肤质感，模拟胶片摄影的皮肤通透感，减少ai感，光影更加出色。
推荐参数设置：
触发词: vifilm
zib训练的lora，zimage base和turbo类模型都可以使用。
turbo权重1.5-2，base权重0.6-0.8
采样器Euler 调度器Simple
想要更明显的胶片颗粒感，可加上提示词film grain，或者调度器选取ddim_uniform，或者降噪强度deniose降低至0.88左右，这三者不建议同时应用，防止画面劣化。
zib训练风格lora不是很容易，因此V1版本有时候在风格生成上尚有不足，我会继续训练和发布后续版本，谢谢大家支持。
提示词使用注意：生图时尽量避免风格类提示词的堆砌、叠加使用，以免影响生成效果，推荐采用自然语言描述。