A simple method of multiple camera calibration for the joint top view projection

Mikhail Mozerov, Ariel Amato, Murad Al Haj, Jordi Gonzàlez

Research output: Contribution to journalArticleResearchpeer-review

3 Citations (Scopus)

Abstract

A simple method for multiple camera calibration based on a novel geometric derivation is presented. The main advantage of this method is that it uses only three points in the world coordinate system to achieve the calibration. Rotation matrix and translation vector for each camera coordinate system are obtained via the given distance between the vertices of the marker triangle formed by the three points. Therefore, the different views from the different cameras can be converted into one top view in the world coordinate system. Eventually, the different trajectories traced by certain tracked agents on the floor plane can be obtained from different viewpoints and can be matched in a joint scene plane. © 2007 Springer-Verlag Berlin Heidelberg.
Original languageEnglish
Pages (from-to)164-170
JournalAdvances in Soft Computing
Volume45
DOIs
Publication statusPublished - 1 Dec 2007

Fingerprint

Dive into the research topics of 'A simple method of multiple camera calibration for the joint top view projection'. Together they form a unique fingerprint.

Cite this