Speeded Up Robust Features
From Scholarpedia
Prof. Luc Van Gool accepted the invitation on 5 February 2010 (self-imposed deadline: 5 August 2010).
Speeded Up Robust Features refers to an interest point detector and descriptor used in computer vision applications. In comparisons, the performance is comparable with that of the seminal SIFT features, but due to the use of simple Haarlet-like wavelets in the derivation, SURF features can be extracted very efficiently. For instance, a 100Hz GPU version was already available in 2009 for video-sized images.
