Hand Gesture Recognition in Video Sequences Using Deep Convolutional and Recurrent Neural Networks

Abstract

Deep learning is a new branch of machine learning,which is widely used by researchers in a lot of artificial intelligence applications,including signal processing and computer vision. The presentresearch investigates the use of deeplearning to solve the hand   gesture   recognition   (HGR)   problem   and   proposes   two models   using   deep   learning   architecture.   The   first   model comprises a convolutional neural network (CNN) and a recurrent neural  network  with a long  short-term  memory  (RNN-LSTM). The  accuracy  of  model  achieves  up  to  82%  when  fed  by  colour channel, and 89% when fed by depth channel. The second model comprises two parallel convolutional neural networks, which aremergedby a merge layer, and a recurrent neural network with a longshort-term memory fed by RGB-D. The accuracy of the latest model achieves up to 93%.

---