scholarly journals British Sign Language Recognition via Late Fusion of Computer Vision and Leap Motion with Transfer Learning to American Sign Language

Sensors ◽  
2020 ◽  
Vol 20 (18) ◽  
pp. 5151
Author(s):  
Jordan J. Bird ◽  
Anikó Ekárt ◽  
Diego R. Faria
Keyword(s):  
Neural Network ◽  
American Sign Language ◽  
Sign Language ◽  
Transfer Learning ◽  
Leap Motion ◽  
Language Recognition ◽  
Sign Language Recognition ◽  
British Sign Language ◽  
Multimodality Approach ◽  
British Sign

In this work, we show that a late fusion approach to multimodality in sign language recognition improves the overall ability of the model in comparison to the singular approaches of image classification (88.14%) and Leap Motion data classification (72.73%). With a large synchronous dataset of 18 BSL gestures collected from multiple subjects, two deep neural networks are benchmarked and compared to derive a best topology for each. The Vision model is implemented by a Convolutional Neural Network and optimised Artificial Neural Network, and the Leap Motion model is implemented by an evolutionary search of Artificial Neural Network topology. Next, the two best networks are fused for synchronised processing, which results in a better overall result (94.44%) as complementary features are learnt in addition to the original task. The hypothesis is further supported by application of the three models to a set of completely unseen data where a multimodality approach achieves the best results relative to the single sensor method. When transfer learning with the weights trained via British Sign Language, all three models outperform standard random weight distribution when classifying American Sign Language (ASL), and the best model overall for ASL classification was the transfer learning multimodality approach, which scored 82.55% accuracy.

scholarly journals British Sign Language Recognition via Late Fusion of Computer Vision and Leap Motion with Transfer Learning to American Sign Language

2020 ◽  
Author(s):  
Jordan J. Bird ◽  
Aniko Ekart ◽  
Diego R. Faria
Keyword(s):  
Computer Vision ◽  
American Sign Language ◽  
Sign Language ◽  
Transfer Learning ◽  
Leap Motion ◽  
Language Recognition ◽  
Late Fusion ◽  
Sign Language Recognition ◽  
Unseen Data ◽  
Modality Approach

In this work, we show that a late fusion approach to multi-modality in sign language recognition improves the overall ability of the model in comparison to the singular approaches of Computer Vision (88.14%) and Leap Motion data classification (72.73%). With a large synchronous dataset of 18 BSL gestures collected from multiple subjects, two deep neural networks are benchmarked and compared to derive a best topology for each. The Vision model is implemented by a CNN and optimised MLP and the Leap Motion model is implemented by an evolutionary optimised deep MLP topology search. Next, the two best networks are fused for synchronised processing which results in a better overall result (94.44%) since complementary features are learnt in addition to the original task. The hypothesis is further supported by application of the three models to a set of completely unseen data where a multi-modality approach achieves the best results relative to the single sensor method. When transfer learning with the weights trained via BSL, all three models outperform standard random weight distribution when classifying ASL, and the best model overall for ASL classification was the transfer learning multi-modality approach which scored 82.55% accuracy.

scholarly journals American Sign Language Recognition Using Leap Motion Controller with Machine Learning Approach

Sensors ◽  
2018 ◽  
Vol 18 (10) ◽  
pp. 3554 ◽  
Author(s):  
Teak-Wei Chong ◽  
Boon-Giin Lee
Keyword(s):  
American Sign Language ◽  
Sign Language ◽  
Recognition System ◽  
Support Vector ◽  
Leap Motion ◽  
American Sign ◽  
Language Recognition ◽  
Motion Controller ◽  
Sign Language Recognition ◽  
Leap Motion Controller

Sign language is intentionally designed to allow deaf and dumb communities to convey messages and to connect with society. Unfortunately, learning and practicing sign language is not common among society; hence, this study developed a sign language recognition prototype using the Leap Motion Controller (LMC). Many existing studies have proposed methods for incomplete sign language recognition, whereas this study aimed for full American Sign Language (ASL) recognition, which consists of 26 letters and 10 digits. Most of the ASL letters are static (no movement), but certain ASL letters are dynamic (they require certain movements). Thus, this study also aimed to extract features from finger and hand motions to differentiate between the static and dynamic gestures. The experimental results revealed that the sign language recognition rates for the 26 letters using a support vector machine (SVM) and a deep neural network (DNN) are 80.30% and 93.81%, respectively. Meanwhile, the recognition rates for a combination of 26 letters and 10 digits are slightly lower, approximately 72.79% for the SVM and 88.79% for the DNN. As a result, the sign language recognition system has great potential for reducing the gap between deaf and dumb communities and others. The proposed prototype could also serve as an interpreter for the deaf and dumb in everyday life in service sectors, such as at the bank or post office.

scholarly journals A Real-Time American Sign Language Recognition System using Convolutional Neural Network for Real Datasets

TEM Journal ◽  
2020 ◽  
pp. 937-943
Author(s):  
Rasha Amer Kadhim ◽  
Muntadher Khamees
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
American Sign Language ◽  
Real Time ◽  
Sign Language ◽  
Recognition System ◽  
High Accuracy ◽  
American Sign ◽  
Language Recognition ◽  
Sign Language Recognition

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.

Sign in / Sign up

Export Citation Format

Share Document