Update README.md
Browse files
README.md
CHANGED
|
@@ -43,11 +43,6 @@ by [DGSpitzer](https://www.youtube.com/channel/UCzzsYBF4qwtMwJaPJZ5SuPg)
|
|
| 43 |
训练用的数据集大部分来自INFLAMES TOYS 1/6神勇关云长兵人模型。因此使用初始词条生成的角色图片会非常接近手办本身（该模型请勿商用，仅作学习研究目的使用），可以通过加入其他的prompt词条，后置主词条的方式来弱化该模型过拟合的问题。
|
| 44 |
|
| 45 |
|
| 46 |
-
|
| 47 |
-
<center><img src="https://huggingface.co/DGSpitzer/Guan-Yu-Diffusion/resolve/main/img/2.jpg" width="768" height="600"/></center>
|
| 48 |
-
|
| 49 |
-
<center><img src="https://huggingface.co/DGSpitzer/Guan-Yu-Diffusion/resolve/main/img/1.jpg" width="768" height="768"/></center>
|
| 50 |
-
|
| 51 |
### 🧨 Diffusers
|
| 52 |
|
| 53 |
This repo contains both .ckpt and Diffuser model files. It's compatible to be used as any Stable Diffusion model, using standard [Stable Diffusion Pipelines](https://huggingface.co/docs/diffusers/api/pipelines/stable_diffusion).
|
|
@@ -80,7 +75,44 @@ You can try the Online Web UI demo build with [Gradio](https://github.com/gradio
|
|
| 80 |
*Buy me a coffee if you like this project ;P ♥*
|
| 81 |
[](https://www.buymeacoffee.com/dgspitzer)
|
| 82 |
|
|
|
|
|
|
|
|
|
|
|
|
|
| 83 |
# **👇Model👇**
|
| 84 |
|
| 85 |
AI Model Weights available at huggingface: https://huggingface.co/DGSpitzer/Guan-Yu-Diffusion/
|
| 86 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 43 |
训练用的数据集大部分来自INFLAMES TOYS 1/6神勇关云长兵人模型。因此使用初始词条生成的角色图片会非常接近手办本身（该模型请勿商用，仅作学习研究目的使用），可以通过加入其他的prompt词条，后置主词条的方式来弱化该模型过拟合的问题。
|
| 44 |
|
| 45 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 46 |
### 🧨 Diffusers
|
| 47 |
|
| 48 |
This repo contains both .ckpt and Diffuser model files. It's compatible to be used as any Stable Diffusion model, using standard [Stable Diffusion Pipelines](https://huggingface.co/docs/diffusers/api/pipelines/stable_diffusion).
|
| 75 |
*Buy me a coffee if you like this project ;P ♥*
|
| 76 |
[](https://www.buymeacoffee.com/dgspitzer)
|
| 77 |
|
| 78 |
+
<center><img src="https://huggingface.co/DGSpitzer/Guan-Yu-Diffusion/resolve/main/img/2.jpg" width="768" height="600"/></center>
|
| 79 |
+
|
| 80 |
+
<center><img src="https://huggingface.co/DGSpitzer/Guan-Yu-Diffusion/resolve/main/img/1.jpg" width="768" height="768"/></center>
|
| 81 |
+
|
| 82 |
# **👇Model👇**
|
| 83 |
|
| 84 |
AI Model Weights available at huggingface: https://huggingface.co/DGSpitzer/Guan-Yu-Diffusion/
|
| 85 |
|
| 86 |
+
# Usage
|
| 87 |
+
|
| 88 |
+
After model loaded, use keyword **guanyu** in your prompt.
|
| 89 |
+
|
| 90 |
+
For sampler, use **Euler A** for the best result (**DDIM** kinda works too), CFG Scale 7, steps 20 should be fine
|
| 91 |
+
|
| 92 |
+
**Example 1:**
|
| 93 |
+
|
| 94 |
+
```
|
| 95 |
+
Portrait of guanyu walking in ancient battlefield, close up shot
|
| 96 |
+
```
|
| 97 |
+
|
| 98 |
+
You can merge the Guan Yu's outfit with other famous characters:
|
| 99 |
+
|
| 100 |
+
**Example 2:**
|
| 101 |
+
|
| 102 |
+
```
|
| 103 |
+
batman, portrait of fancy superhero guanyu, golden spiderman, mech, robot, high tech, shining core, intricate details, 4k
|
| 104 |
+
```
|
| 105 |
+
|
| 106 |
+
**Example 3 (Additional keywork Wuxia):**
|
| 107 |
+
|
| 108 |
+
There is another hidden keyword you can use in this model **wuxia**, to generate traditional Chinese martial art character:
|
| 109 |
+
|
| 110 |
+
```
|
| 111 |
+
a gorgeous wuxia girl standing in the palace
|
| 112 |
+
```
|
| 113 |
+
|
| 114 |
+
---
|
| 115 |
+
|
| 116 |
+
**NOTE: usage of this model implies accpetance of stable diffusion's [CreativeML Open RAIL-M license](LICENSE)**
|
| 117 |
+
|
| 118 |
+
---
|