This is the pytorch implement of our paper "RSBuilding: Towards General Remote Sensing Image Building Extraction and Change Detection with Foundation Model"
I read in your paper that "For the building extraction task, which includes building segmentation annotations, we create a duplicate of I1, denoted as I2, and apply random photometric distortion."
But I can't find where exactly in your code this is done. Could you give me some pointers?
I read in your paper that "For the building extraction task, which includes building segmentation annotations, we create a duplicate of I1, denoted as I2, and apply random photometric distortion." But I can't find where exactly in your code this is done. Could you give me some pointers?