Enforcing non-positive weights for stable support vector tracking

In this paper we demonstrate that the support vector tracking (SVT) framework first proposed by Avidan is equivalent to the canonical Lucas-Kanade (LK) algorithm with a weighted Euclidean norm. From this equivalence we empirically demonstrate that in many circumstances the canonical SVT approach is unstable, and characterize these circumstances theoretically. We then propose a novel "nonpositive support kernel machine" (NSKM) to circumvent this limitation and allow the effective use of discriminative classification within the weighted LK framework. This approach ensures that the pseudo-Hessian realized within the weighted LK algorithm is positive semidefinite which allows for fast convergence and accurate alignment/tracking. A further benefit of our proposed method is that the NSKM solution results in a much sparser kernel machine than the canonical SVM leading to sizeable computational savings and much improved alignment performance.

• Type: Conference
• Year: 2008
• Where: CVPR
• Authors: Simon Lucey
• Keywords: Canonical Svm, Canonical Svt Approach, Computer Vision, CVPR 2008, Weighted Euclidean Norm, Weighted Lk Algorithm, Weighted Lk Framework