Sciweavers

Share
DAGM
2007
Springer

Greedy-Based Design of Sparse Two-Stage SVMs for Fast Classification

9 years 1 months ago
Greedy-Based Design of Sparse Two-Stage SVMs for Fast Classification
Cascades of classifiers constitute an important architecture for fast object detection. While boosting of simple (weak) classifiers provides an established framework, the design of similar architectures with more powerful (strong) classifiers has become the subject of current research. In this paper, we focus on greedy strategies recently proposed in the literature that allow to learn sparse Support Vector Machines (SVMs) without the need to train full SVMs beforehand. We show (i) that asymmetric data sets that are typical for object detection scenarios can be successfully handled, and (ii) that the complementary training of two sparse SVMs leads to sequential two-stage classifiers that slightly outperform a full SVM, but only need about 10% kernel evaluations for classifying a pattern.
Rezaul Karim, Martin Bergtholdt, Jörg H. Kapp
Added 14 Aug 2010
Updated 14 Aug 2010
Type Conference
Year 2007
Where DAGM
Authors Rezaul Karim, Martin Bergtholdt, Jörg H. Kappes, Christoph Schnörr
Comments (0)
books