Speeded-Up Robust Features (SURF) is a feature detection algorithm used in many [tag:computer-vision] tasks, such as object recognition, 3D reconstruction, camera calibration, and [tag:cbir].
Speeded-Up Robust Features (SURF) is an algorithm in computer vision to detect and describe local features in images.
SURF is billed as specifically “Speeded-Up” and “Robust” over its predecessor, the SIFT algorithm – it is based on Hessian blob detection, and eschews the more byzantine arrangements of SIFT (e.g. octaves). This visualization illustrates SURF features with circles, scaled accordingly for each feature, overlaid on the image from which they were extracted:
The standard standard SURF implementation, from the introductory journal article, is about an order of magnitude faster than David Lowe’s implementation of SIFT (when run in a comparable environment).