På svenska

Presentation Information     2010-05-03 (15:15)   •  The seminar room at Vi2

Speaker Khalid Niazi
Title A brief tutorial on Scale-Invariant Feature Transform
Abstract Scale-invariant Feature Transform (SIFT) is a well known algorithm to extract reliable image features which are invariant to scale, illumination, rotation and location. SIFT can be divided into four main parts, i.e., scale-space extrema detection, keypoint localization, orientation assignment and keypoint descriptor construction. It has been widely used in image registration, face recognition, motion tracking, panorama image construction, etc. I will present a brief tutorial on SIFT, addressing both its theoretical and implementation aspects.