-
Notifications
You must be signed in to change notification settings - Fork 121
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
some fixes #30
base: master
Are you sure you want to change the base?
some fixes #30
Conversation
python 3.8 has dependency conflicts under Ubuntu, python 3.9 does not.
Improve error handling when images cannot be loaded
def process(self, img_path): | ||
img = cv2.imread(img_path) | ||
if img is None: | ||
print(f"Error: Unable to load image '{img_path}'. Skipping...") | ||
return None |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I wonder if the change here is necessary because it will also affect ImageColorizationPipelineHF class.
class ImageColorizationPipelineHF(ImageColorizationPipeline): |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
My intention, not yet complete, was to modify the pipeline to process a directory of images and save them, rather than display the results. But I defer to you, of course. I think this project has a lot less color bleeding than others I've tried, but some of my test images are over-saturated. I'll probably add some color adjustment parameters.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Hi @gessyoo, thank you for your recognition of our project!
I understand what you're trying to do, but I think it's better to keep the process() method the same behavior as before. As for processing image directories and saving them, maybe we can write these codes outside of the class, or to write a separate class method (such as process_img_folder() or something).
We do observe that some results can be oversaturated, so adding some color adjustment parameters is a good idea!
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I've added some experimental color correction options with defaults to produce a hopefully more natural, less saturated output. I've also updated the teaser.png to show the color correction effect.
added color adjustment options to improve output colors, reduce over-saturation
Updated teaser.png to show experimental color adjustment options applied to test images
minor modifications to overcome some install and inference issues.