ATDAN Style LoRA

中文 | English 使用前阅读 | Read before use

—

仿ATDAN的作画风格的LoRA模型。

我对肢体拟合已经进行了较大幅度的修正，但是目前肢体拟合度仍然过高，一般情况下成图虽然没问题，至少示例图的复杂度是没问题的，但在复杂姿势下的生成会有局限性，我会在未来找到方法后处理。

权重建议值：1 (or 0.6~1.2)

基于模型：Anything-5RE

训练信息：共计263张原画（含裁剪图），最后一次训练迭代1038步。

拟合比率：

• 肢体拟合：23%（这导致部分复杂姿势会无法生成，自行斟酌其局限性）

• 人物拟合：30%

• 构图拟合：35%

• 技法拟合：130%

其他的一些建议：EasyNegative, badhandv4。

LoRA model with a painting style imitating ATDAN.

I have made significant adjustments to the limb fitting, but currently, the degree of limb fitting is still too high. Although there is generally no problem with creating images and at least the complexity of sample images is not a problem, there are limitations in generating complex poses. I will work on finding a solution for this issue in the future.

Recommended weight value: 1 (or ranging from 0.6 to 1.2).

Based on the model: Anything-5RE

Training information: A total of 263 paintings (including cropped images) were used for training, with the final training iteration completed at 1,038 steps.

Fitting ratio:

• Motion fitting: 23%（This results in some complex postures that cannot be generated, consider their limitations）

• Character fitting: 30%

• Composition fitting: 35%

• Technique fitting: 130%

Other suggestions: EasyNegative, badhandv4.

—

题外话和互动环节

请访问：[V3 Final]Artist: mignon | Mignon artist’s painting style – v3.0 | Stable Diffusion LoRA | Civitai

Digression & Interaction

Please visit: [V3 Final]Artist: mignon | Mignon artist’s painting style – v3.0 | Stable Diffusion LoRA | Civitai

—

任何问题，欢迎反馈，问题不提出就不会得到解决。如果你创作出了好作品，欢迎分享它。如果它无法满足你的需求，我是说，我的朋友，请不要给出一星好评。

Any questions are welcome for feedback, as problems will not be solved without being raised. If you have created a great work, please feel free to share it. If it cannot meet your needs, my friend, please do not give a one-star rating.