Model-Based Person Tracking Through Networked Cameras

Authors: Kyoung-Mi Lee, Youn-Mi Lee

Abstract:

This paper proposes a way to track persons by making use of multiple non-overlapping cameras. Tracking persons on multiple non-overlapping cameras enables data communication among cameras through the network connection between a camera and a computer, while at the same time transferring human feature data captured by a camera to another camera that is connected via the network. To track persons with a camera and send the tracking data to another camera, the proposed system uses a hierarchical human model that comprises a head, a torso, and legs. The feature data of the person being modeled are transferred to the server, after which the server sends the feature data of the human model to the cameras connected over the network. This enables a camera that captures a person's movement entering its vision to keep tracking the recognized person with the use of the feature data transferred from the server.

Keywords: Person tracking, human model, networked cameras, vision-based surveillance.

Digital Object Identifier (DOI): doi.org/10.5281/zenodo.1070587

References:

[1] S. Park and J. K. Aggarwa, "Segmentation and tracking of interacting human body parts under occlusion and shadowing," in Proc. of International workshop on motion and video computing, pp.105-111, 2002.
[2] Javed, Z. Rasheed, O. Alatas and M. Shah, "KNIGHTM: A real-time surveillance system for multiple overlapping and non-overlapping cameras," in Proc. of the International conference on multimedia and expo, 2003.
[3] F.M.Porikli and A. Divakaran, "Multi-camera calibration, object tracking and query generation", in Proc. of the International conference on multimedia and expo, pp. 653-656, 2003.
[4] Y. Matsushita, K. Nishino, K. Ikeuchi and M. Sakauchi, "Illumination normalization with time-dependent intrinsic images for video surveillance," IEEE trans. on PAMI, 26(10):1336-1347, 2004.
[5] S. Khan and M. Shah, "Consistent labeling of tracked objects in multiple cameras with overlapping Fields of View," IEEE trans. on PAMI, 25(10):1355-1360, 2003.
[6] K. M. Lee and Y. M. Lee, "Tracking multi -person robust to illumination changes and occlusions," in Proc. of International of conference on artificial reality and telexistence, pp. 429-432, 2004.