Input: is the tracking result of prior frame, is set of
candidate samples credible positions of the center-point coordinates in next
frame. is Over-complete dictionary, is frame numbers.
(1) for
(2) SRGC
  calculate =
(3) calculate identity ; where ,
(4) obtaining the center point from
(5) obtaining the by and Affine transformation of
(6) SRLC to ;
  calculate ;
(7) calculate identity , where
Output:
Algorithm 1: Algorithm of Fisher discrimination dictionary learning.