Abstract
We present an application of the FuzzyBoost learning algorithm, where the weak learners select spatio-temporal groups of features for waving detection. The features encode the spatial distribution of the optic flow of a tracked person, considering the polar sampling of the flow for each instant. The FuzzyBoost algorithm selects groups of features that discriminate better than any single feature, bringing robustness and generalization over the TemporalBoost algorithm. © 2013 Springer-Verlag.
Cite
CITATION STYLE
Moreno, P., & Santos-Victor, J. (2013). Waving detection using the FuzzyBoost algorithm and flow-based features. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 7950 LNCS, pp. 19–26). https://doi.org/10.1007/978-3-642-39094-4_3
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.
