-
Notifications
You must be signed in to change notification settings - Fork 366
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[WIP] Colorization + ControlNet #27
Comments
Hi, really nice results, I am training ControlNet (with SD2.1) on images from COCO and noticed similar issues. In particular, here is what I found:
Despite this, the overall quality is still incredible and these artifacts are understandable in the end, as it is a generative model and it is basically "inventing" the colors. I was thinking of customizing the text prompt used in training using the color tonality of the image, e.g., extracting the overall tonality of the image (cold, warm, bright, dark) and embedding this information in the prompt. If this gives some improvements, I will post in this discussion |
@rensortino @tg-bomze Hey guys, I am also working on gray image colorization with sd + controlnet. But I'm wondering how to set the prompts during training? In ColorizeNet(https://github.com/rensortino/ColorizeNet), the author used 10 prompts (basically all variants of “colorize this image”) and randomly provide them during training along with the gray image. Do you guys have any other suggestions? Any advices will be appriciated! Thanks! |
could not be easier :) i made something similar here: https://huggingface.co/GeroldMeisinger/controlnet-channels (colorization from missing RGB channel). btw there are already some colorization controlnet in CN 1.1 and the first colorization CN ever here https://civitai.com/models/80549/color-based-picture-control-controlnet |
Hello, we have been trying (at neural.love) to train the colorization model based on the ControlNet architecture.
The model was trained on different LR's on the manually collected b&w-colorful image pares dataset.
Any recommendations regarding the training process are highly appreciated.
Current model problems:
It is still heavily work-in-progress, but it could already (sometimes) colorize quite well, which is why we decided to share it. I have created a pull request here, and here are some results of the first version:
![01](https://user-images.githubusercontent.com/48222107/232609449-21a99c2a-2708-4433-9acc-2361eb207237.jpg)
![02](https://user-images.githubusercontent.com/48222107/232609587-633a0eba-0f31-46dd-86ae-5255b7827e36.jpg)
![03](https://user-images.githubusercontent.com/48222107/232609613-50caec31-aebf-4d7a-9a3c-460972200162.jpg)
![04](https://user-images.githubusercontent.com/48222107/232609627-f16003c3-d968-4e64-aaa4-1ab755aa6c6e.jpg)
![05](https://user-images.githubusercontent.com/48222107/232609649-a2d1ef4e-40c1-4f87-a090-c241baba820a.jpg)
![06](https://user-images.githubusercontent.com/48222107/232609667-90dfc61d-c23d-4019-a7df-d5398ce3b8ca.jpg)
![07](https://user-images.githubusercontent.com/48222107/232609696-ba802267-6e4f-4f0f-aeec-3f1ccd597ac4.jpg)
![08](https://user-images.githubusercontent.com/48222107/232609722-3ba3676a-b036-4389-b346-1ded60d5f2df.jpg)
The text was updated successfully, but these errors were encountered: