Real Time Hand Gesture Recognition in Depth Image using CNN

Abstract

Hand gestures can play a notable role in computer vision, and hand gesture-based methods can stand out in providing a native way of interaction. Deafness is a degree of loss such that a person is unable to understand speech, spoken language. Sign language declined the gap in spoken language. The hand gesture is analyzed identically to sign language presenting the naturalness of intercommunication for deaf people. Real-time hand gesture recognition has been proposed in our research. Our proposed CNN model architecture will remediate the communication barrier of deaf people. The proposed model has achieved an accuracy of 94.61% to recognize 11 several different gestures using depth images.