|M.Sc Student||Kaplan Avi|
|Subject||Interpreting the Ratio Criterion for Matching SIFT|
|Department||Department of Computer Science||Supervisors||Professor Michael Lindenbaum|
|Dr. Tamar Avraham|
|Full Thesis text|
Matching keypoints by minimizing the Euclidean distance between their SIFT descriptors is an effective and extremely popular technique. Using the ratio between distances, as suggested by Lowe, is even more effective and leads to excellent matching accuracy. Probabilistic approaches, modeling the distribution of the distances, were found effective as well.
This work focuses, for the first time, on analyzing Lowe's ratio criterion using a probabilistic approach. We provide two alternative interpretations of this criterion, which show that it is not only an effective heuristic but can also be formally justified. Our first interpretation justifies Lowe's ratio by showing that it corresponds to a conditional probability that the match is incorrect. Our second interpretation shows that the ratio corresponds to the Markov bound on this probability. The interpretation enables us to make the ratio criterion even more effective, and the obtained matching performance exceeds all previous (non-learning based) results.