The decomposition method is currently one of the major methods for solving the convex quadratic optimization problems being associated with support vector machines. For a special c...
This paper describes recognition of emotions of an unkown person during natural walking. As gait data is redundant, high dimensional and variable, effective feature extraction is ...
— We propose a method for automatic emotion recognition as part of the FERA 2011 competition [1] . The system extracts pyramid of histogram of gradients (PHOG) and local phase qu...
Abhinav Dhall, Akshay Asthana, Roland Goecke, Tom ...
In this paper, we present a Support Vector Machine (SVM) based ensemble approach to combat the extractive multi-document summarization problem. Although SVM can have a good general...
We present a novel pedestrian detection system based on probabilistic component assembly. A part-based model is proposed which uses three parts consisting of head-shoulder, torso a...
Martin Rapus, Stefan Munder, Gregory Baratoff, Joa...