Innovation in learning - The use of avatar for sign language

Abstract

This paper presents the steps followed in developing an avatar-interpreter of the Brazilian sign language for deaf (LIBRAS), applied to an electro technical glossary. The research was done in collaboration between The Surface Interaction and Displays Division (DSID) and The National Service for Industrial Apprenticeship (SENAI "Ítalo Bologna"), a reference center in the attendance of people having physical or mental incapability. This work makes use of advanced techniques of motion capture, treatment of images and virtualization, to produce an avatar that mimics a teacher-interpreter of the specific electro technical signs of LIBRAS during the lesson. The technology used in this work is a VICON system with 8 cameras that emit and capture infrared light, and the open source tools Blender and Make-Human. © 2013 Springer-Verlag.