Static Hand Gesture Recognition for American Sign Language using Deep Convolutional Neural Network

Prangon Das, Tanvir Ahmed, Md. Firoj Ali

20202020 IEEE Region 10 Symposium (TENSYMP)59 citationsDOI

Abstract

One of the complicated issues of computer vision is the recognition of any sign languages. The deaf and the dumb people use these sign languages to communicate. In the area of deep learning at recent progress there are numerous applications that neural networks can have for interpreting sign languages. The recognition of American Sign Language static images employing a capable artificial intelligence tool, Convolutional Neural Network has been proposed in this paper. ASL dataset of 1815 images of 26 English alphabets has been used to train and validate our model. Validation accuracy has been found 94.34% which is better than many existing methods.

Topics & Concepts

Computer scienceConvolutional neural networkSign languageAmerican Sign LanguageGestureSign (mathematics)Gesture recognitionArtificial intelligenceDeep learningArtificial neural networkSpeech recognitionNatural language processingLinguisticsMathematical analysisMathematicsPhilosophyHand Gesture Recognition SystemsGait Recognition and AnalysisHuman Pose and Action Recognition