In this paper, we propose a human action recognition system suitable for embedded computer vision applications in security systems, human-computer interaction and intelligent envi...
Abstract. Score functions induced by generative models extract fixeddimensions feature vectors from different-length data observations by subsuming the process of data generation, ...
Alessandro Perina, Marco Cristani, Umberto Castell...
A fundamental problem in peer-to-peer networks is how to locate appropriate peers efficiently to answer a specific query request. This paper proposes a model in which semantically...
The work investigates the use of multi dimensional histograms for segmentation of images of chronic wounds. We employ a Support Vector Machine (SVM) classifier for automatic extrac...
Recent works in object recognition often use visual words, i.e. vector quantized local descriptors extracted from the images. In this paper we present a novel method to build such ...