SignQuiz: A Quiz Based Tool for Learning Fingerspelled Signs in Indian Sign Language Using ASLR

Joy, Jestin; Balakrishnan, Kalpana; Sreeraj, M

doi:10.1109/access.2019.2901863

Cited by 40 publications

(17 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In [14], American sign language alphabet recognition was developed by extracting features from depth images using PCANet. Joy et al [20] developed a quiz-based tool to learn finger spelled signs for Indian sign language. Arabic language recognition became a hot topic nowadays [21].…”

Section: Related Workmentioning

confidence: 99%

DeepArSLR: A Novel Signer-Independent Deep Learning Framework for Isolated Arabic Sign Language Gestures Recognition

Aly

2020

IEEE Access

View full text Add to dashboard Cite

Hand gesture recognition has attracted the attention of many researchers due to its wide applications in robotics, games, virtual reality, sign language and human-computer interaction. Sign language is a structured form of hand gestures and the most effective communication way among hear-impaired people. Developing an efficient sign language recognition system to recognize dynamic isolated gestures encounters three major challenges, namely, hand segmentation, hand shape feature representation and gesture sequence recognition. Traditional sign language recognition methods utilize color-based hand segmentation algorithms to segment hands, hand-crafted feature extraction for hand shape representation and Hidden Markov Model (HMM) for sequence recognition. In this paper, a novel framework is proposed for signerindependent sign language recognition using multiple deep learning architectures comprising hand semantic segmentation, hand shape feature representation and deep recurrent neural network. The recently developed semantic segmentation method called DeepLabv3+ is trained using a set of pixel-labeled hand images to extract hand regions from each frame of the input video. Then, the extracted hand regions are cropped and scaled to a fixed size to alleviate hand scale variations. Extracting hand shape features is achieved using a single layer Convolutional Self-Organizing Map (CSOM) instead of relying on transfer learning of pretrained deep convolutional neural networks. The sequence of extracted feature vectors are then recognized using deep Bi-directional Long Short-Term Memory (BiLSTM) recurrent neural network. BiLSTM network contains three BiLSTM layers, one fully connected and softmax layers. The performance of the proposed method is evaluated using a challenging Arabic sign language database containing 23 isolated words captured from three different users. Experimental results show that the performance of proposed framework outperforms with large margin the state-of-the-art methods for signer-independent testing strategy. INDEX TERMS Arabic sign language recognition, deep learning, hand semantic segmentation, convolutional self-organizing map, signer-independent, deep BiLSTM network.

show abstract

Section: Related Workmentioning

confidence: 99%

DeepArSLR: A Novel Signer-Independent Deep Learning Framework for Isolated Arabic Sign Language Gestures Recognition

Aly

2020

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Dour and Sharma [81] a neural-fuzzy fusion is done to reduce complexity of similar gesture recognition. To improve efficiency a multiscale oriented histogram within addition to contour directions is used for feature extraction [9]. This integration of approaches makes system memory efficient with high recognition rate of 97.1%.…”

Section: Hybrid Technique For Feature Extractionmentioning

confidence: 99%

“…Automatic SL recognition starts from data acquisition phase. This can be done in two ways, i.e., computer vision [1][2][3][4][5][6] and sensor enabled device [7][8][9][10]. Camera is used for capturing gestures in formal technique while devices (Kinect sensor, data gloves, flex sensor, LMC) with sequence of sensor are used for capturing images in later method.…”

Section: Introductionmentioning

confidence: 99%

Feature Extraction Technique for Vision-Based Indian Sign Language Recognition System: A Review

Tyagi

Bansal

2020

Computational Methods and Data Engineering

View full text Add to dashboard Cite

Vision-based sign language recognition system is an emerging research area in human-computer interaction (HCI). It has been proved as a powerful communication tool for deaf and mute society, irrespective of geographical differences. Automatic sign language recognition system works broadly in three phases namely image pre-processing, feature extraction and classification of gestures. The output of the feature extraction phase is crucial for various classifiers. This paper presents a comprehensive review of feature extraction techniques used in vision-based sign language recognition system. A taxonomy of currently used techniques for feature extraction has been presented. The paper concludes by presenting future direction in feature extraction technique for Indian sign language (ISL).

show abstract

“…El aprendizaje del lenguaje de señas comienza con el Fingerspelled; de ahí, surge la herramienta SignQuiz, óptima para el aprendizaje de signos Fingerspe, soportada en la web rentable usada en la India (ISL) y la aplicación de reconocimiento automático de este tipo de lenguaje. Esta aplicación ayuda en la ilustración, sin la participación de facilitadores externos; además, con la misma se tiene el propósito de aprender gestos a través de los dedos, empleando redes neuronales (sistemas de neuronas artificiales conectadas para la trasmisión de diversas señales), evidenciando una alternativa positiva para el proceso de aprendizaje (Joy et al 2019).…”

Section: Investigativas Tendencias Investigativas | Revista Virtualunclassified

Tendencias investigativas en el estudio de tecnologías inclusivas para población sorda

Patiño-Toro

Patiño-Vanegas

Toro³

et al. 2020

Rev.virtual univ. catol. norte

View full text Add to dashboard Cite

El avance tecnológico debe orientarse a generar mejoras en la calidad de vida y bienestar de la sociedad. Una de las poblaciones con mayores necesidades corresponde a personas sordas, por esto resulta necesario examinar la evolución y tendencias investigativas, en relación con tecnologías dirigidas a población con dificultades auditivas, entre 1990 y 2019. La metodología involucró 202 documentos recopilados mediante ecuación de búsqueda, obtenida de Scopus, originando indicadores de cantidad y calidad, además del análisis de temas emergentes y crecientes en este campo. Los hallazgos incluyen prevalencia del interés por la exploración y difusión de las tecnologías diseñadas para población sorda, especialmente en países como Estados Unidos, demostrándose altos índices de productividad y circulación del conocimiento por parte de instituciones y autores. Entre las conclusiones, se evidenció la importancia de investigaciones en temáticas asociadas al lenguaje de señas, accesibilidad, reconocimiento de gestos mediante sensores, sustitución sensorial y aplicaciones móviles, como instrumentos favorecedores del desarrollo de personas afectadas en su audición. Se reconoce el valor de las investigaciones y la evolución tecnológica como mecanismos dinamizadores del desarrollo de personas sordas, basadas en la comunicación y los procesos de aprendizaje, convirtiéndose en elemento integrador de esta población en distintos contextos sociales.

show abstract

SignQuiz: A Quiz Based Tool for Learning Fingerspelled Signs in Indian Sign Language Using ASLR

Cited by 40 publications

References 29 publications

DeepArSLR: A Novel Signer-Independent Deep Learning Framework for Isolated Arabic Sign Language Gestures Recognition

DeepArSLR: A Novel Signer-Independent Deep Learning Framework for Isolated Arabic Sign Language Gestures Recognition

Feature Extraction Technique for Vision-Based Indian Sign Language Recognition System: A Review

Tendencias investigativas en el estudio de tecnologías inclusivas para población sorda

Contact Info

Product

Resources

About