Closed uhSuiL closed 2 months ago
This project is currently only applicable to text images with single-line text, and the input size is limited to 128*512 patches. You can detect the text line image from the original image through the PaddleOCR or other text detection models, then crop and resize the patches to 128 x 512.
Thanks for your reply and appreciate your work
If so, is there any suggested strategy to process img of abitrary size (like 300x300)?