%0 Thesis %1 magalhaes2018reconhecimento %A Magalhães, Gabriel Ilharco %C São José dos Campos, São Paulo, Brazil %D 2018 %I Instituto Tecnológico de Aeronáutica (ITA) %K brazilian-sign-language bsl computer-vision deep-learning gesture-recognition libras machine-learning neural-network real %T Reconhecimento de Gestos da Língua Brasileira de Sinais Através de Redes Neurais %U http://gabrielilharco.com/publications/reconhecimento_de_gestos_da_lingua_brasileira_de_sinais_atraves_de_redes_neurais.pdf %X In a world where more than 70 million people rely on sign language to communicate,a system capable of recognizing and translating gestures to written or spoken language has great social impact. Despite rights claimed in recent decades, the deaf community still faces many challenges due to communication barriers. Gesture recognition, crucial for translation, is an active research topic in the computer vision and machine learning communities, and has been studied for decades. Among the most common approaches for this task, there are the electronic gloves with sensors, depth camera based approaches and simple camera based approaches. This last method has the advantage of completeness,since in many sign languages, including Brazilian Sign Language, other parts of the body such as the face and its expressions are needed to recognize some gestures. Additionally,it relies only on commonly found technologies, in contrast to the other approaches. We present a state-of-the art approach, using a simple color camera for real-time static and continuous recognition of gestures from Brazilian Sign Language. For static recognition, we create a data set with 33000 examples of 30 gestures; for continuous recognition,we create a data set with 2000 videos containing phrases with 72 distinct gestures. Both data sets were built without restrictions with respect to clothing, background, lighting or distance between the camera and the user, commonly found in other studies. We propose end-to-end systems for each case using Deep Learning: for the former, a system based on a Deep Convolutional Residual Neural Network; for the latter, a hybrid architecture using Long-Short Term Memory Cells on top of convolutional layers. Our method shows state of the art accuracy for both cases and is capable of running in real time.

@mastersthesis{magalhaes2018reconhecimento, abstract = {In a world where more than 70 million people rely on sign language to communicate,a system capable of recognizing and translating gestures to written or spoken language has great social impact. Despite rights claimed in recent decades, the deaf community still faces many challenges due to communication barriers. Gesture recognition, crucial for translation, is an active research topic in the computer vision and machine learning communities, and has been studied for decades. Among the most common approaches for this task, there are the electronic gloves with sensors, depth camera based approaches and simple camera based approaches. This last method has the advantage of completeness,since in many sign languages, including Brazilian Sign Language, other parts of the body such as the face and its expressions are needed to recognize some gestures. Additionally,it relies only on commonly found technologies, in contrast to the other approaches. We present a state-of-the art approach, using a simple color camera for real-time static and continuous recognition of gestures from Brazilian Sign Language. For static recognition, we create a data set with 33000 examples of 30 gestures; for continuous recognition,we create a data set with 2000 videos containing phrases with 72 distinct gestures. Both data sets were built without restrictions with respect to clothing, background, lighting or distance between the camera and the user, commonly found in other studies. We propose end-to-end systems for each case using Deep Learning: for the former, a system based on a Deep Convolutional Residual Neural Network; for the latter, a hybrid architecture using Long-Short Term Memory Cells on top of convolutional layers. Our method shows state of the art accuracy for both cases and is capable of running in real time.}, added-at = {2019-09-11T07:30:29.000+0200}, address = {São José dos Campos, São Paulo, Brazil}, author = {Magalhães, Gabriel Ilharco}, biburl = {https://www.bibsonomy.org/bibtex/256c3f1b43bbab1060d5d553d654a0977/jpmor}, interhash = {5ee4431a53b1a8f9122a16f925c5d137}, intrahash = {56c3f1b43bbab1060d5d553d654a0977}, keywords = {brazilian-sign-language bsl computer-vision deep-learning gesture-recognition libras machine-learning neural-network real}, publisher = {Instituto Tecnológico de Aeronáutica (ITA)}, school = {Instituto Tecnológico de Aeronáutica (ITA)}, timestamp = {2020-10-07T13:36:50.000+0200}, title = {Reconhecimento de Gestos da Língua Brasileira de Sinais Através de Redes Neurais}, url = {http://gabrielilharco.com/publications/reconhecimento_de_gestos_da_lingua_brasileira_de_sinais_atraves_de_redes_neurais.pdf}, year = 2018 }