GESTURE-DETECT: A Deep Learning Framework Using Convolution Neural Network for Hand Gesture

Abstract

Hand gesture recognition refers to non-verbal mode of communication used by deaf and mute community to communicate with normal individuals. Human Computer Interaction (HCI) covers somewhat an interface, comprising hand motions also known as Gestures. A framework possibly used to detect hand motions and express data to operate device. This is an important domain in HCI that comprises interface of device and users. Hand motion detection involves recording specific gestures and detecting them using a camera. Hand motions serve multiple communication purposes. It can help those with disabilities, including as hearing and speech difficulties, as well as stroke patients, communicate and meet their fundamental needs. The technique of Human computer interaction is crucial in health sectors as for as the people with hearing loss and have speech disorders. Regardless of substantial advances for hand motion detection in computer vision-based system for language interpretation, in this sector a significant challenge remains which is limited in scope and only considers common motions. As a result, an innovative hand-vision based Deep learning model which is Convolution Neural Network provides various advantages, including increased accuracy, robustness to variations, and real-time performance. This framework can amplify for synchronous (Real-time) functioning, trained from larger datasets, and can be effortlessly scaled to handle complicated recognition tasks ultimately enabling more effective forms of HCI.