Robust estimation of heights of moving people using a single camera

Abstract

In recent years, there has been increased interest in characterizing and extracting 3D information from video sequences for object tracking and identification. In this paper, we propose a single view-based framework for robust estimation of height and position. In this work, 2D features of a target object is back-projected into the 3D scene space where its coordinate system is given by a rectangular marker. Then the position and height are estimated in the 3D scene space. In addition, geometric error caused by an inaccurate projective mapping is corrected by using geometric constraints provided by the marker. The proposed framework is entirely non-iterative, and therefore is very fast. As the proposed framework uses a single camera, it can be directly embedded into conventional monocular camera-based surveillance/security systems. The accuracy and robustness of the proposed technique are verified on the experimental results of several real video sequences taken from outdoor environments.