Update README.md
Browse files
README.md
CHANGED
@@ -16,8 +16,9 @@ Controlnet SDXL Tile model realistic version, fit for both webui extention and c
|
|
16 |
|
17 |
<!-- Provide a longer summary of what this model is. -->
|
18 |
- This is a SDXL based controlnet Tile model, trained with huggingface diffusers sets, fit for Stable diffusion SDXL controlnet.
|
19 |
-
- It is
|
20 |
-
- As there is no SDXL Tile
|
|
|
21 |
|
22 |
![image/png](https://cdn-uploads.huggingface.co/production/uploads/641edd91eefe94aff6de024c/ddFT3326ddNOWBeoFnfZl.png)
|
23 |
![image/png](https://cdn-uploads.huggingface.co/production/uploads/641edd91eefe94aff6de024c/OETMPhSCVEKdyUvILMsyp.png)
|
@@ -33,15 +34,15 @@ Controlnet SDXL Tile model realistic version, fit for both webui extention and c
|
|
33 |
|
34 |
|
35 |
## Uses
|
36 |
-
- **Important: Tile model is not a upscale model!!! it enhance or change the detial of the
|
37 |
-
- This model will not significant change the base model style. it only adding the
|
38 |
- --Just use a regular controlnet model in Webui by select as tile model and use tile_resample for Ultimate Upscale script.
|
39 |
- --Just use load controlnet model in comfyui and apply to control net condition.
|
40 |
- --if you try to use it in webui t2i, need proper prompt setup, otherwise it will significant modify the original image color. I don't know the reason, as I don't really use this function.
|
41 |
- --it do perform much better with the image from the datasets. However, everything works fine for the i2i model and what is the place usually the ultimate upscale is applied!!
|
42 |
- **--Please also notice this is a realistic training set, so no comic, animation application are promised.**
|
43 |
- --For tile upscale, set the denoise around 0.3-0.4 to get good result.
|
44 |
-
- --For controlnet
|
45 |
- --For human image fix, IPA and early stop on controlnet will provide better reslut
|
46 |
- **--Pickup a good realistic base model is important!**
|
47 |
![image/jpeg](https://cdn-uploads.huggingface.co/production/uploads/641edd91eefe94aff6de024c/zPyYn2fSFmD1Q07ME0Hkg.jpeg)
|
|
|
16 |
|
17 |
<!-- Provide a longer summary of what this model is. -->
|
18 |
- This is a SDXL based controlnet Tile model, trained with huggingface diffusers sets, fit for Stable diffusion SDXL controlnet.
|
19 |
+
- It is original trained for my personal realistic model project used for Ultimate upscale process to boost the picture details. with a proper workflow, it can provide a good result for high detailed, high resolution image fix.
|
20 |
+
- As there is no SDXL Tile available from the most open source, I decide to share this one out.
|
21 |
+
- I will share my workflow soon as I am still working on it to have better result.
|
22 |
|
23 |
![image/png](https://cdn-uploads.huggingface.co/production/uploads/641edd91eefe94aff6de024c/ddFT3326ddNOWBeoFnfZl.png)
|
24 |
![image/png](https://cdn-uploads.huggingface.co/production/uploads/641edd91eefe94aff6de024c/OETMPhSCVEKdyUvILMsyp.png)
|
|
|
34 |
|
35 |
|
36 |
## Uses
|
37 |
+
- **Important: Tile model is not a upscale model!!! it enhance or change the detial of the original size image, remember this before you use it!**
|
38 |
+
- This model will not significant change the base model style. it only adding the features to the upscaled pixel blocks....
|
39 |
- --Just use a regular controlnet model in Webui by select as tile model and use tile_resample for Ultimate Upscale script.
|
40 |
- --Just use load controlnet model in comfyui and apply to control net condition.
|
41 |
- --if you try to use it in webui t2i, need proper prompt setup, otherwise it will significant modify the original image color. I don't know the reason, as I don't really use this function.
|
42 |
- --it do perform much better with the image from the datasets. However, everything works fine for the i2i model and what is the place usually the ultimate upscale is applied!!
|
43 |
- **--Please also notice this is a realistic training set, so no comic, animation application are promised.**
|
44 |
- --For tile upscale, set the denoise around 0.3-0.4 to get good result.
|
45 |
+
- --For controlnet strength, set to 0.9 will be better choice
|
46 |
- --For human image fix, IPA and early stop on controlnet will provide better reslut
|
47 |
- **--Pickup a good realistic base model is important!**
|
48 |
![image/jpeg](https://cdn-uploads.huggingface.co/production/uploads/641edd91eefe94aff6de024c/zPyYn2fSFmD1Q07ME0Hkg.jpeg)
|