Live demonstration: A HMM-based real-time sign language recognition system with multiple depth sensors

Author(s):  
Kai-Yin Fok ◽  
Chi-Tsun Cheng ◽  
Nuwan Ganganath
Author(s):  
Zhibo Wang ◽  
Tengda Zhao ◽  
Jinxin Ma ◽  
Hongkai Chen ◽  
Kaixin Liu ◽  
...  

Sign language is the only method of communication for the hearing and speech impaired people around the world. Most of the speech and hearing-impaired people understand single sign language. Thus, there is an increasing demand for sign language interpreters. For regular people learning sign language is difficult, and for speech and hearing-impaired person, learning spoken language is impossible. There is a lot of research being done in the domain of automatic sign language recognition. Different methods such as, computer vision, data glove, depth sensors can be used to train a computer to interpret sign language. The interpretation is being done from sign to text, text to sign, speech to sign and sign to speech. Different countries use different sign languages, the signers of different sign languages are unable to communicate with each other. Analyzing the characteristic features of gestures provides insights about the sign language, some common features in sign languages gestures will help in designing a sign language recognition system. This type of system will help in reducing the communication gap between sign language users and spoken language users.


TEM Journal ◽  
2020 ◽  
pp. 937-943
Author(s):  
Rasha Amer Kadhim ◽  
Muntadher Khamees

In this paper, a real-time ASL recognition system was built with a ConvNet algorithm using real colouring images from a PC camera. The model is the first ASL recognition model to categorize a total of 26 letters, including (J & Z), with two new classes for space and delete, which was explored with new datasets. It was built to contain a wide diversity of attributes like different lightings, skin tones, backgrounds, and a wide variety of situations. The experimental results achieved a high accuracy of about 98.53% for the training and 98.84% for the validation. As well, the system displayed a high accuracy for all the datasets when new test data, which had not been used in the training, were introduced.


Measurement ◽  
2021 ◽  
Vol 168 ◽  
pp. 108431
Author(s):  
M.A. Ahmed ◽  
B.B. Zaidan ◽  
A.A. Zaidan ◽  
Mahmood M. Salih ◽  
Z.T. Al-qaysi ◽  
...  

Sign in / Sign up

Export Citation Format

Share Document