Action and gesture recognition provides important information for interaction between human and devices that monitors living, healthcare facilities or entertainment activities in smart homes. Recent years, there are many learning machine models studying to recognize human action and gesture. In this paper, we propose a dynamic hand gesture recognition system based on two stream-convolution network (ConvNet) architecture. Besides, we also modify the method to enhance its performance that is suitable for indoor application. Our contribution is improvement of two stream ConvNet to achieve better performance. We use MobileNet-V2 as an extractor since it has less number of parameters and volume than other convolution networks. The results show that the proposal model improves execution speed and memory resource usage comparing to existing models. | Two-stream convolutional network for dynamic hand gesture recognition using convolutional long short-term memory networks
