Gesture Recognition System Based on Neural Networks by Using COTS RFID Tag Array
Jiaying Wu, Chuyu Wang, Lei Xie · 2020
Nowadays, gesture recognition plays a more and more important role in human-computer interaction. In this regard, contact sensors or computer vision have made some progress, but they also have shortcomings in portability or privacy. In this work, we propose a gesture recognition system which uses RFID tag array and neural networks to recognize gestures. By using an RFID tag array, we can obtain gesture information in a non-contact, non-infringing manner. By combining CNN and LSTM as CNN-LSTM, we can focus on both spatial and temporal features and get better performance. Experiments show that the accuracy of the system on the test set is 92.17%, and it performs well in recognizing different gestures of different users at different speeds.