Mean Shift: theory and applications

Speaker:  Marco Cristani - Università di Verona
  Tuesday, July 4, 2006 at 5:30 PM caffè, tè & C. ore 17.00
The Mean Shift procedure is an old non-parametric density estimation technique, that has been recently re-considered and partially re-invented by numerous authors. The goals that can be achieved with this technique are various, ranging from image segmentation to multidimensional data clustering. In this seminar, the basis of the Mean Shift paradigm are given, starting from standard notions of statistical pattern recognition, such as Parzen Windows and Kernel-based methods. Moreover, novel applications and strategies developed in our department are presented, which show how the Mean Shift paradigm can be applied to 3D data segmentation problems and medical data imaging tasks, in an efficient and effective way.

Ca' Vignal - Piramide, Floor 0, Hall Verde

Programme Director

External reference
Publication date
June 12, 2006