Implementação de técnicas de visão computacional aplicadas no ensino de expressões em Libras: um estudo de caso IFPI - Campus Floriano

Abstract

This study aimed to provide a didactic approach to teaching Brazilian Sign Language (Libras) with the goal of enhancing teaching-learning strategies and fostering the inclusion and accessibility of deaf individuals in society. This was accomplished through the development of a computational tool employing computer vision techniques to recognize Libras expressions. Addressing these challenges, numerous studies have explored the use of computer vision techniques to aid in Libras education, leveraging image processing and pattern recognition to analyze and interpret sign language gestures. The process was divided into four stages: data collection, model training, model testing, and interface construction. Following the development and validation of the tool, it was found capable of recognizing the majority of gestures introduced during training. This project not only achieved its primary objective but also paved the way for future research and implementations at the intersection of technology and sign language education.