Classification of Desk Workers’ Behaviors for a Service Robot

Currently, many people are working their works at a desk at an office or their homes, and work is the one problem that makes humans stressed. A service robot is the one type of robot created to interact with a customer, and the service robot can be a friend while we are working. It can know a customer’s behavior and make an action intimately, and it will help us reduce the stress. Therefore, this research aims at customer behavior classification while working at a desk for a service robot. This research method consists of 3 steps that are 1) skeleton feature extraction, 2) data processing, and 3) behavior classification. We have first gotten the human skeleton information on an input RGB image using the OpenPose framework. Then, we have calculated the vectors and angles of the skeleton information to be the input features of a model. Finally, we used the input features from the previous step to classify human behavior using the Deep Learning algorithm. This research method can predict customer behavior with the F1 score of each behavior is above 0.95.