Explicit-Visual-Prompt
Explicit-Visual-Prompt copied to clipboard
about the Defocus Blur Detection CUHK datasets
appreciate to your great work.
when I would like to retrain your model on dataset of Defocus Blur Detection in CUHK, I can't get the 604 images of training dataset and 100 images of testing dataset.
the dataset link you give is 1000 images and GTs,why in your paper is 604-train imags and 100-test images? how can I get your splited image datasets?