3D Arabic sign language recognition using linear combination of multiple 2D views
Mohamed F. Tolba, Ahmed Samir, Magdy Aboul-Ela · International Conference on Informatics and Systems · 2012
Earlier researchers in sign language recognition faced a problem in some signs because of the single view based recognition. A model is proposed and developed for multiple-views hand postures recognition. Pulse Coupled Neural Network is used to generate features vector for single view. Two views with different view angles are used; each view generates its features vector. The two 2D-vectors then are linearly combined with weights to produce 3D features which will be used in recognition