Fast Real-Time Pipeline for Robust Arm Gesture Recognition

This paper presents a real-time pipeline for dynamic arm gesture recognition based on OpenPose keypoint estimation, keypoint normalization, and a recurrent neural network classifier.The 1 1 normalization scheme and two feature representations (coordinateand angle-based) are presented for the pipeline.In addition, an efficient method to improve robustness against camera angle variations is also introduced by using artificially rotated training data.Experiments on a custom traffic-control gesture dataset demonstrate high accuracy across varying viewing angles and speeds.Finally, an approach to calculate the speed of the arm signal (if necessary) is also presented.