Verified: a year ago
SafeTensor
The FLUX.1 [dev] Model is licensed by Black Forest Labs. Inc. under the FLUX.1 [dev] Non-Commercial License. Copyright Black Forest Labs. Inc.
IN NO EVENT SHALL BLACK FOREST LABS, INC. BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH USE OF THIS MODEL.
This is a lora used to generate Chinese paper-cut style images. As one of China's intangible cultural heritage, paper-cutting has a distinctive artistic style, mainly in red, with simple patterns and strong symbolic significance, and is often used to express themes such as flowers, animals and people. By fine-tuning the specific style on the basis of FLUX model, this LoRA enhances the expressive force of the model in generating paper-cut style images.
Description:
Training data: Select representative paper-cut photo collections from all over China.
Generation effect: The LoRA can generate images with strong Chinese paper-cut artistic characteristics, with clear edges, concise and vivid shapes and strong folk art atmosphere.
Optimization and adaptation: The LoRA has made special color optimization for the red paper-cut style during data training, but it also supports the paper-cut style application of other colors. Users can adjust the color matching and style details of the generated image according to their needs.
Usage: Please use [Chinese paper cutting] as the trigger word. After the trigger word is turned on, try it from about 0.6, and adjust it according to the specific effect.
Because of the special artistic style of paper-cutting, it is recommended to use concise natural language as a hint, and generally there is obvious effect between 0.5 and 1; If a long text with a large number of words collapses, you can increase the weight to more than 1, and then adjust it according to the situation. You can usually try between 1 and 1.5.
You can use [red lines] [yellow lines] [blue lines] [black and white line] to adjust the color of paper-cutting, and red is recommended.
For example, it begins with: Chinese paper cutting, yellow lines, ...
——————————————————
这是一个用于生成中国剪纸风格图像的lora。剪纸作为中国非物质文化遗产之一，具有鲜明的艺术风格，主要以红色为主，图案简单且具备极强的象征意义，常用于表现花卉、动物、人物等主题。该LoRA通过在FLUX模型的基础上进行特定风格的微调，增强了模型在生成剪纸风格图像方面的表现力。
说明：
训练数据：从中国各地挑选具备代表性的剪纸图片集。
生成效果：该LoRA能够生成具有强烈中国风剪纸艺术特色的图像，作品边缘清晰，造型简洁而生动，表现出浓厚的民间艺术氛围。
优化与适配：该LoRA在数据训练时针对红色的剪纸风进行了特别的颜色优化，但也支持其他颜色的剪纸风格应用，用户可根据需求调节生成图像的配色与风格细节。
使用：请使用【Chinese paper cutting】作为触发词，开启触发词后，一般从0.6左右开始尝试，根据具体效果进行调节。
由于剪纸特殊的艺术风格，推荐使用简洁的自然语言作为提示词，0.5-1之间一般就有明显的效果；如果使用字数较多的长文本后出现崩坏，可以提高权重到1以上，再视情况调整。1-1.5之间通常都可以尝试。
可以使用【Red lines】【 yellow lines】【blue lines】【Black and white line】来调整剪纸的颜色，推荐红色。
比如这样开头：Chinese paper cutting， yellow lines，……