Hi, thanks for great work on this dataset. You mentioned in your paper that due to the large size of the images you applied a cropping with overlaps strategy to convert an original image to multiple small images with size 1024x2048. Is there a version of this dataset that can be downloaded or maybe you can share a script to create this version of the dataset?
Hi, thanks for great work on this dataset. You mentioned in your paper that due to the large size of the images you applied a cropping with overlaps strategy to convert an original image to multiple small images with size 1024x2048. Is there a version of this dataset that can be downloaded or maybe you can share a script to create this version of the dataset?