I'm diving into the world of OpenCv and there are some concept very used but that lack of definition: For example: What are exactly a keypoint, a descriptor and extractor.
I'm guessing the definition by their names but I'd appreciate a brief explanation of each.
Thank you very much.
A keypoint is a pixel in the image for which you want to extract some information. It can either be a random position in the image or a position that has some kind of property, more of that at the extractor part.
A keypoint extractor is an automated technique for extracting keypoints in a given image that have some kind of property. For example a strong edge response, a specific color, a high response to a specific Gabor filter, ...
Finally a keypoint descriptor takes a region around the found keypoint and tries to come up with a compact description of that region. This is used for later comparison, learning, ... The possibilities here are also endless.
But you should benefit from reading a general computer vision book, something like http://szeliski.org/Book/
Asked: 2017-03-22 06:09:36 -0600
Seen: 231 times
Last updated: Mar 22 '17
