Beforehand, I must apologize, I am just a beginner in image processing and I am not very familiar with many computer vision operations. Here is my problem:
I am using deep learning to perform a segmentation task as a first step. From this segmentation, I want to find coordinates for later use. In order to save memory and time, I was wondering if it was possible to first downsample the image, process it through the neural network, find the coordinates, and then by mean of calculation I presume, obtain the corresponding coordinates on the original image (not downsample) ?
Does anyone have an idea ?