Action recognition using hybrid spatio-temporal bag-of-features

In this paper we addresses the problem of human action recognition by introducing a new representation of image sequences as a collection of spatiotemporal events that are localized at interest point and using multi-class SVM for classification. The interest points are detected by the SIFT detector and a spatio-temporal interest point detector. We proposed a new bag of words approach to represent videos in two different model. A multi-class SVM scheme that is based on one-class hypersphere SVM is used for classification. We also present action classification results on two different datasets. Our results are comparable to previous published results on these datasets.