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 E...
This paper reports our knowledge-ignorant machine learning approach to the triage task in TREC2004 genomics track, which is actually a text categorization problem. We applied Supp...
Kernel-based methods, e.g., support vector machine (SVM), produce high classification performances. However, the computation becomes time-consuming as the number of the vectors su...