Image and Video Analysis class final project, by Michele Polese, Davide Magrin and Andrea Dittadi. In this project we introduced two descriptors that can be used for hand gesture recognition by means of machine learning techniques, and we provided some performance analysis of such descriptors. We used a support vector machine (SVM) as classifier, training it with pairs consisting of some image descriptor and the corresponding desired output class. After the training phase, the SVM was used to classify new hand gestures.
For more details, the report can be found here.