We know that the image pyramid is used to detect largely invariant feature points in the ORB algorithm. How to display matched feature points in the image pyramid reduced in size from the original image in the original image? Can I simply upscale and display it?
