Optical flow-based person tracking by multiple cameras

Abstract

This paper describes optical flow-based person tracking using multiple cameras in an indoor environment. There are usually several objects in indoor environments which may obstruct a camera view. If we use only one camera, tracking may fail when the target is occluded by the objects. By using multiple cameras, this problem can be solved. In our method, each camera tracks the target person independently. By exchanging information among cameras, the three dimensional position and the velocity of the target are estimated. When a camera loses the target by occlusion, the target position and the velocity in the image are estimated using information from other cameras which are tracking the target.

A Multi-Robot System Architecture for Trajectory Control of Groups of People

“…For the subfigure 12-(c), circles were plotted with radius of 3σ d . Similar works with distributed networked systems for multi-people tracking but different contexts were presented by (Nakazawa et al, 1998;Tsutsui et al, 2001). …”

Principal axis-based correspondence between multiple cameras for people tracking

“…In [6,7], object centroids are taken as feature points and correspondence is established by estimating the corresponding 3D centroids in the world coordinate system. In [8], all cameras are calibrated and the 3D environment model is known beforehand.…”

3d Trajectory Reconstruction of a Moving Object From a Stereo Video Using Particle Filter

“…Tsutsui et al [7] uses stereovision and optical flow to recreate the 3D trajectory of a moving person. The optical flow method is very sensitive to illumination change and require static background model.…”