Radar-based Dataset Development for Human Activity Recognition

Human activity recognition (HAR) is an important research problem and considerable efforts have been made in the past two decades to develop successful solutions. Datasets developed for HAR serve as a baseline for evaluating new HAR algorithms and play a pivotal role in advancing the research efforts in this domain. Hundreds of datasets have been developed by various researchers which focus on different types of motions and activities [1]. In this context, most of the efforts have been invested in developing the datasets which contain either the video or clips of 2D images of different types of motions. Due to the advent of 3D sensors, modern datasets also contain the information of the motion as well as the depth [2].