SkygirlLuna

I'm wondering what is changes in the reimplementation compared to https://github.com/clovaai/CRAFT-pytorch

input image size

1

What is the maximal image input size for craft text detection? Is the success rate for text detection dependent on the image size?

When running python ic15/script.py -g=gt.zip -s=pixellinkch4.zip or the other 2 commands python curved_tiou/script.py -g=ctw1500-gt.zip -s=det_ctw1500.zip python curved_tiou/script.py -g=total-text-gt.zip -s=total-text_baseline.zip I get an error: Traceback (most recent call last): File "curved_tiou/script.py",...