Add thanks section in README file

README.md

@@ -12,9 +12,9 @@ tags:
 
 # SDXL ControlNet: DWPose
 
-Here are controlnet weights trained on stabilityai/stable-diffusion-xl-base-1.0 with [DWPose](https://github.com/IDEA-Research/DWPose) conditioning.
+Here are ControlNet weights trained on stabilityai/stable-diffusion-xl-base-1.0 with [DWPose](https://github.com/IDEA-Research/DWPose) conditioning.
 
-### Using in 🧨 diffusers
+## Using in 🧨 diffusers
 
 First, install all the libraries:
 
@@ -23,7 +23,7 @@ pip install -q easy-dwpose transformers accelerate
 pip install -q git+https://github.com/huggingface/diffusers
 ```
 
-#### Example 1
+### Example 1
 
 To generate a realistic DJ with the following image driving the pose:
 
@@ -87,7 +87,7 @@ Image generated by SDXL is:
 
 ![Pose 1](./images/dwpose_image_1.png)
 
-#### Example 2
+### Example 2
 
 To generate a anime version of a woman sitting on a bench with the following image driving the pose:
 
@@ -153,7 +153,7 @@ Image generated by SDXL is:
 
 ![Pose 2](./images/dwpose_image_2.png)
 
-### Training
+## Training
 
 The [training script](https://github.com/huggingface/diffusers/blob/main/examples/controlnet/README_sdxl.md) by HF🤗 was used. 
 
@@ -170,4 +170,14 @@ Data parallel with a single GPU batch size of 2 with gradient accumulation 8.
 Constant learning rate of 8e-5
 
 #### Mixed precision
-fp16
+fp16
+
+## Thanks
+
+- [raulc0399](https://huggingface.co/raulc0399): for highly inspiring me with the creation of the [DWpose dataset](https://huggingface.co/datasets/dimitribarbot/dw_pose_controlnet) based on the [Openpose one](https://huggingface.co/datasets/raulc0399/open_pose_controlnet).
+
+- [thibaud](https://huggingface.co/thibaud): for highly inspiring me with the hyper parameters of the HF training script, based on the [Openpose ControlNet](https://huggingface.co/thibaud/controlnet-openpose-sdxl-1.0).
+
+- [RedHash](https://huggingface.co/RedHash): for the [easy_dwpose](https://github.com/reallyigor/easy_dwpose) module, which highly simplifies the DWPose inference and which I used in the examples above.
+
+- [Hugging Face](https://huggingface.co/): for the ControlNet training script 🤗.