A comprehensive evaluation of deep models and optimizers for Indian sign language recognition

Sharma, Prachi; Anand, R. S.

doi:10.1016/j.gvc.2021.200032

Cited by 24 publications

(16 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This includes acknowledging that a gap needs to be eliminated between speech-impaired community members and the rest of the community, as pointed out by [ 24 , 47 ]. The success of developing a deep learning model that could predict sign language at such a high level of accuracy, precision, and reliability confirms the findings of other researchers [ 27 ] that the deep learning model can be relied upon to narrow this gap [ 27 ]. Although the actual experiment conducted in this research did not compare efficiency between conduct-based human action recognition and vision-based human action recognition, this research agrees with their argument that the latter could be preferable.…”

Section: Discussionsupporting

confidence: 76%

“…Educational institutions can use this technology to facilitate learning activities and interactions between hearing-impaired and visually impaired students [ 26 ]. In addition, sign language and human action recognition have been applied in computer and mobile phones, gaming interfaces, machine control, robotics, and televisions [ 27 ]. Sign language and human action recognition can be applied for easy interaction between children and computers, development of recognizable forensic identification, and video conferencing communication [ 28 ].…”

Section: Literature Reviewmentioning

confidence: 99%

“…The results obtained from the experiment showed a 97% signer-independent recognition rate and outperformed the pre-existing models by the accurate specification of the hand and heads positions, respectively. Deep learning CNN has nowadays gained much popularity in data science research, including the application of vision-based hand gesture recognition for sign language interpretation using deep learning [ 27 ]. In one study, a Deep learning-based CNN that was specifically designed for sign language recognition was used.…”

Section: Literature Reviewmentioning

confidence: 99%

“…Likewise, Ref. [ 27 ] developed a sign language recognition model using CNN to translate Indian sign Language to text. This research was termed the first comprehensive analysis of Indian sign language.…”

Section: Literature Reviewmentioning

confidence: 99%

“…Different researchers have provided various recommendations for further research into CNN-related models. However, another challenging aspect lies in determining the best model to adopt and improve, as the interdependence of different models with respect to the optimization of tuning parameters must be considered [ 27 ]. From the above literature review, it is evident that most of the CNN models to date have been developed for Indian, Arabic, and Chinese sign languages, among others.…”

Section: Literature Reviewmentioning

confidence: 99%

See 4 more Smart Citations

Sign and Human Action Detection Using Deep Learning

2022

View full text Add to dashboard Cite

Human beings usually rely on communication to express their feeling and ideas and to solve disputes among themselves. A major component required for effective communication is language. Language can occur in different forms, including written symbols, gestures, and vocalizations. It is usually essential for all of the communicating parties to be fully conversant with a common language. However, to date this has not been the case between speech-impaired people who use sign language and people who use spoken languages. A number of different studies have pointed out a significant gaps between these two groups which can limit the ease of communication. Therefore, this study aims to develop an efficient deep learning model that can be used to predict British sign language in an attempt to narrow this communication gap between speech-impaired and non-speech-impaired people in the community. Two models were developed in this research, CNN and LSTM, and their performance was evaluated using a multi-class confusion matrix. The CNN model emerged with the highest performance, attaining training and testing accuracies of 98.8% and 97.4%, respectively. In addition, the model achieved average weighted precession and recall of 97% and 96%, respectively. On the other hand, the LSTM model’s performance was quite poor, with the maximum training and testing performance accuracies achieved being 49.4% and 48.7%, respectively. Our research concluded that the CNN model was the best for recognizing and determining British sign language.

show abstract

Section: Discussionsupporting

confidence: 76%