Update README.md
Browse files
README.md
CHANGED
@@ -10,195 +10,5 @@ pipeline_tag: text-to-image
|
|
10 |
|
11 |
# **ControlNet++: All-in-one ControlNet for image generations and editing!**
|
12 |
## **ProMax Model has released!! 12 control + 5 advanced editing, just try it!!!**
|
13 |
-
|
14 |
-
|
15 |
-
## Network Arichitecture
|
16 |
-

|
17 |
-
|
18 |
-
## Advantages about the model
|
19 |
-
- Use bucket training like novelai, can generate high resolutions images of any aspect ratio
|
20 |
-
- Use large amount of high quality data(over 10000000 images), the dataset covers a diversity of situation
|
21 |
-
- Use re-captioned prompt like DALLE.3, use CogVLM to generate detailed description, good prompt following ability
|
22 |
-
- Use many useful tricks during training. Including but not limited to date augmentation, mutiple loss, multi resolution
|
23 |
-
- Use almost the same parameter compared with original ControlNet. No obvious increase in network parameter or computation.
|
24 |
-
- Support 10+ control conditions, no obvious performance drop on any single condition compared with training independently
|
25 |
-
- Support multi condition generation, condition fusion is learned during training. No need to set hyperparameter or design prompts.
|
26 |
-
- Compatible with other opensource SDXL models, such as BluePencilXL, CounterfeitXL. Compatible with other Lora models.
|
27 |
-
|
28 |
-
|
29 |
-
***We design a new architecture that can support 10+ control types in condition text-to-image generation and can generate high resolution images visually comparable with
|
30 |
-
midjourney***. The network is based on the original ControlNet architecture, we propose two new modules to: 1 Extend the original ControlNet to support different image
|
31 |
-
conditions using the same network parameter. 2 Support multiple conditions input without increasing computation offload, which is especially important for designers
|
32 |
-
who want to edit image in detail, different conditions use the same condition encoder, without adding extra computations or parameters. We do thoroughly experiments
|
33 |
-
on SDXL and achieve superior performance both in control ability and aesthetic score. We release the method and the model to the open source community to make everyone
|
34 |
-
can enjoy it.
|
35 |
-
|
36 |
-
Inference scripts and more details can found: https://github.com/xinsir6/ControlNetPlus/tree/main
|
37 |
-
|
38 |
-
**If you find it useful, please give me a star, thank you very much**
|
39 |
-
|
40 |
-
**~500+ star, release the ProMax version with tile and inpainting!!!~**
|
41 |
-
**SDXL ProMax version has been released!!!,Enjoy it!!!**
|
42 |
-
**1000+ star, release the ControlNet++ model for SD3!!!**
|
43 |
-
**3000+ star, release the ControlNet++ ProMax model for SD3!!!**
|
44 |
-
|
45 |
-
Note: we put the promax model with a promax suffix in the same [huggingface model repo](https://huggingface.co/xinsir/controlnet-union-sdxl-1.0), detailed instructions will be added later.
|
46 |
-
## Advanced editing features in Promax Model
|
47 |
-
### Tile Deblur
|
48 |
-

|
49 |
-

|
50 |
-

|
51 |
-

|
52 |
-

|
53 |
-

|
54 |
-
### Tile variation
|
55 |
-

|
56 |
-

|
57 |
-

|
58 |
-

|
59 |
-

|
60 |
-

|
61 |
-
|
62 |
-
### Tile Super Resolution
|
63 |
-
Following example show from 1M resolution --> 9M resolution
|
64 |
-
<div style="display: flex; justify-content: space-between;">
|
65 |
-
<img src="./images/tile_super1.webp" alt="Image 1" style="width: 49%; margin: 1%;">
|
66 |
-
<img src="./images/tile_super1_9upscale.webp" alt="Image 2" style="width: 49%; margin: 1%;">
|
67 |
-
</div>
|
68 |
-
|
69 |
-
<div style="display: flex; justify-content: space-between;">
|
70 |
-
<img src="./images/tile_super2.webp" alt="Image 1" style="width: 49%; margin: 1%;">
|
71 |
-
<img src="./images/tile_super2_9upscale.webp" alt="Image 2" style="width: 49%; margin: 1%;">
|
72 |
-
</div>
|
73 |
-
|
74 |
-
### Image Inpainting
|
75 |
-

|
76 |
-

|
77 |
-

|
78 |
-

|
79 |
-

|
80 |
-

|
81 |
-
|
82 |
-
### Image Outpainting
|
83 |
-

|
84 |
-

|
85 |
-

|
86 |
-

|
87 |
-

|
88 |
-

|
89 |
-
|
90 |
-
|
91 |
-
## Visual Examples
|
92 |
-
### Openpose
|
93 |
-

|
94 |
-

|
95 |
-

|
96 |
-

|
97 |
-

|
98 |
-
### Depth
|
99 |
-

|
100 |
-

|
101 |
-

|
102 |
-

|
103 |
-

|
104 |
-
### Canny
|
105 |
-

|
106 |
-

|
107 |
-

|
108 |
-

|
109 |
-

|
110 |
-
### Lineart
|
111 |
-

|
112 |
-

|
113 |
-

|
114 |
-

|
115 |
-

|
116 |
-
### AnimeLineart
|
117 |
-

|
118 |
-

|
119 |
-

|
120 |
-

|
121 |
-

|
122 |
-
### Mlsd
|
123 |
-

|
124 |
-

|
125 |
-

|
126 |
-

|
127 |
-

|
128 |
-
### Scribble
|
129 |
-

|
130 |
-

|
131 |
-

|
132 |
-

|
133 |
-

|
134 |
-
### Hed
|
135 |
-

|
136 |
-

|
137 |
-

|
138 |
-

|
139 |
-

|
140 |
-
### Pidi(Softedge)
|
141 |
-

|
142 |
-

|
143 |
-

|
144 |
-

|
145 |
-

|
146 |
-
### Teed
|
147 |
-

|
148 |
-

|
149 |
-

|
150 |
-

|
151 |
-

|
152 |
-
### Segment
|
153 |
-

|
154 |
-

|
155 |
-

|
156 |
-

|
157 |
-

|
158 |
-
### Normal
|
159 |
-

|
160 |
-

|
161 |
-

|
162 |
-

|
163 |
-

|
164 |
-
|
165 |
-
## Multi Control Visual Examples
|
166 |
-
### Openpose + Canny
|
167 |
-

|
168 |
-

|
169 |
-

|
170 |
-

|
171 |
-

|
172 |
-

|
173 |
-
|
174 |
-
### Openpose + Depth
|
175 |
-

|
176 |
-

|
177 |
-

|
178 |
-

|
179 |
-

|
180 |
-

|
181 |
-
|
182 |
-
### Openpose + Scribble
|
183 |
-

|
184 |
-

|
185 |
-

|
186 |
-

|
187 |
-

|
188 |
-

|
189 |
-
|
190 |
-
### Openpose + Normal
|
191 |
-

|
192 |
-

|
193 |
-

|
194 |
-

|
195 |
-

|
196 |
-

|
197 |
-
|
198 |
-
### Openpose + Segment
|
199 |
-

|
200 |
-

|
201 |
-

|
202 |
-

|
203 |
-

|
204 |
-

|
|
|
10 |
|
11 |
# **ControlNet++: All-in-one ControlNet for image generations and editing!**
|
12 |
## **ProMax Model has released!! 12 control + 5 advanced editing, just try it!!!**
|
13 |
+
This is just a copy of the original repo [here](https://huggingface.co/xinsir/controlnet-union-sdxl-1.0) by xinsir.
|
14 |
+
The only difference is I have renamed the Promax model to make it easier to use.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|