The  Weizmann  Institute  of  Science
                  Faculty of Mathematics and Computer Science


                          Vision and Robotics Seminar

                                  Shai Avidan
                            Interdisciplinary Center


                                 will speak on


                              Subset Selection for
                  Computationally Efficient SVM Classification

Abstract:
We update the {\bf SVM} score of an object through a video sequence with a small and variable
subset of support vectors. In the first frame we use all the support vectors to compute the {\bf SVM}
score of the object but in subsequent frames we use only a small and variable subset of support
vectors to update the {\bf SVM} score. In each frame we calculate the dot-products of the support
vectors in the subset with the pattern of the object in the current frame. The difference in the
dot-products, between past and current frames, is used to update the {\bf SVM} score. This is done
at a fraction of the computational cost required to re-evaluate the {\bf SVM} score from scratch in
every frame. The two methods we develop are ``loop unrolling", in which we break the set of all
support vectors into subsets of equal size and use them cyclically, and ``Maximum Variance Subset
Selection'', in which we choose the support vectors whose dot-product with the test pattern varied
the most in previous frames. We combine this techniques together for the problem of maintaining the
{\bf SVM} score of objects through a video sequence. Results on real video sequences are shown.





                      The lecture will take place in the 
                     Lecture Hall, Room 1, Ziskind Building
                          on Thursday, January 9, 2003
                                    at 11:00