Service Robot for Deaf and Speech-impaired People
Published in Robotics, National Taiwan University, 2018
Trained a convolutional neuron network to recognize sign language videos. Implemented the system to a service humanoid robot to assist deaf and speech-impaired people.
Abstract
In this project, we built an application for Pepper which can serve deaf and speech-impaired people. Deep learning is used to recognize sign language so that Pepper can understand and interact with those who cannot speak and hear. SLAM is also implemented to construct a map for the surroundings and for further navigation. With the help of this application, Pepper is able to guide or provide other services for those disabled people.