Person following based on Haar-like feature and HOG feature in indoor environment

Abstract

In order to follow people in indoor environment, an approach for person detection and following is proposed based on Haar-like feature and Histogram of Oriented Gradients (HOG) feature for a service robot. The Haar-like feature detector based on Adaptive Boost (Adaboost) Cascade algorithm is applied to detect the human face when human faces the service robot, in which the HOG feature detector based on Support Vector Machine (SVM) algorithm is used to detect and follow human side-by-side or back to the camera. Several experiments are implemented with PR2 robot to verify the feasibility and effectiveness of the proposed method. Experimental results show this method has high accuracy and precision in detecting person.