Inquiry About Using Non-Square Images for ControlNet Training

Subject: Inquiry About Using Non-Square Images for ControlNet Training

Dear [Team/Developer],

First of all, thank you for providing the training code for ControlNet. I have recently been utilizing this code to train the ControlNet for an sdxl model, and I am quite satisfied with the results. However, I believe there is room for improvement.

During the training process, I noticed that the program crops images into squares. The images I am using are from interior design, which are rarely square and tend to lose their integrity when cropped. I am wondering if it is possible to train with longer or wider images, such as those with aspect ratios of 3:2 or 2:3. Would it be sufficient to modify the image processing part of the code to accommodate these dimensions?

I look forward to your response and thank you very much for your support!

Best regards

Start by locating the ControlNet/SDXL training code's image-processing entry point and tracing where input images are cropped into squares. Check how 3:2 and 2:3 inputs are batched and represented during training; done means those aspect ratios can be trained without destructive square cropping and the existing training checks still pass.

Vom Indexierungsmodell aus dem Issue-Text verfasst.