I am trying to develop a program for matching video requests with video in a database. So something like a Google image search, but for videos. I am using OpenCV for this implementation.
I can successfully generate function descriptors for selected frames in the video (which tbh is not so complicated).
I do not know how I should implement part of the database. I was wondering if anyone could give me instructions on what already exists in OpenCV to facilitate my implementation: what class extensions, how to structure them, any existing implementations / examples, ...?
Basically, I would have a set of descriptors from the request clip, which I would like to compare with a large (large) database with pre-computed descriptors, and then using the voting algorithm we will return the best match.
So, I'm not asking what algorithms to use, I just want to know about the best practices in OpenCV to implement the kind of thing I'm describing.
I am using OpenCV 2.4.8 C ++ on Mac in Xcode, if that matters
EDIT
Let me ask my question a little more specifically. If I use the SIFT SURF detector / descriptor, I get a lot of features. Doing this for a bunch of frames from multiple videos will lead to a lot of data.
FlannMatcher ( - ), ... ... - . , , ...