Randomized Trees and Ferns: Keypoint Matching by Classification

Before BRIEF, we developed a fast method for keypoint recognition based on classification. We first used Randomized Trees as the classifier, and then developed the Ferns, a simplified version relying on a Naive Bayesian approach for better performances.

Videos


click on each image to download the video

Test Sequences

Original frames (259Mb !).

Relevant Publications

M. Calonder; V. Lepetit; P. Fua; K. Konolige; J. Bowman et al. : Compact Signatures for High-speed Interest Point Description and Matching. 2009. IEEE International Conference on Computer Vision, Kyoto, JAPAN, Sep 29-Oct 02, 2009. p. 357-364.
M. Calonder; V. Lepetit; P. Fua : Keypoint Signatures for Fast Learning and Recognition. 2008. European Conference on Computer Vision , Marseille, FRANCE, Oct 12-18, 2008. p. 58-71.
M. Özuysal; M. Calonder; V. Lepetit; P. Fua : Fast Keypoint Recognition using Random Ferns; IEEE Transactions on Pattern Analysis and Machine Intelligence. 2010. DOI : 10.1109/TPAMI.2009.23.
V. Lepetit; P. Fua : Keypoint Recognition using Randomized Trees; Transactions on Pattern Analysis and Machine Intelligence. 2006.