SURF - Speeded Up Robust Features

SURF stands for Speeded-Up Robust Features. It is inspired by SIFT, and is intended for applications that requires speed but does not require the precision of SIFT. SURF generates 64 features compared to SIFT which generates 128 features.

• SURF notes by Yet Another Blogger here.

FLANN Overview

Fast Library Approximate Nearest Neighbor (FLANN) algorithm was proposed by Marius Muja and David G. Lowe.

FLANN is a way to solve the nearest neighbor search (NNS), also known as proximity search, similarity search or closest point search, is an optimization problem for finding closest points in metric spaces. Donald Knuth in vol. 3 of The Art of Computer Programming (1973) called it the post-office problem, referring to an application of assigning to a residence the nearest post office.

A simple way to solve the nearest neighbor problem is by the k-NN algorithm, but k-NN will not work in very complex problems such as matching SIFT or SURF keypoints. An intermediate solution is to use KD-trees.

• Webpage by the original authors is here
• Available in OpenCV, which demo code here.
  
• FLANN notes by Yet Another Blogger here.

SIFT and RANSAC

http://web.engr.oregonstate.edu/~hess/index.html

http://en.wikipedia.org/wiki/RANSAC