Hand Gesture Recognition for Emoji Prediction
Jatin Gupta · International Journal for Research in Applied Science and Engineering Technology · 2020
Emojis are ideograms and smileys visual symbols that are used widely in wireless communication. They present rich novel possibilities of representation and interaction as a new modality. They exist in various genres, including hand gestures, human faces, figures, and signs. Hand gestures, which are the most common and intuitive non-verbal means of communication when we are using a computer, and related work, has recently sparked an interest. Hands often appear in images, videos, and their appearances and pose can give important clues about what people are doing. A combination of hand gestures and emojis can communicate and express the message very conveniently. Considering the positive outcomes of image recognition from specific deep learning methods, we suggest an emoji predictor in real-time. This project consists of a hand gesture recognition method and emoji generator using filters to detect hands and Convolutional Neural Network (CNN) for training the model. Here, a database is being created of hand gestures to train the system. The prediction will be focused on the hand movements and capture the changes by preparing for different hand movement positions to the highest degree of precision.