Open adnankhalid7454 opened 1 year ago
Hi, Sorry for the late response, the ROI extraction is done by applying a region detection algorithm like YOLO, and then applying the same coordinates found, on the ground truth mask images. If you don't have that first object detection step or you don't plan to apply it, you will have to manually prepare the data for the segmentation
Can please provide the code for the preparation of CBIS-DDSM dataset. (I-e, Extraction of ROI from Input and Ground annotation). The actual dataset has no information about the ROI; it only has annotation files for the mass region.