In this paper, we address the problem of recovering the intrinsic and extrinsic parameters of a camera or a group of cameras in a setting overlooking a traffic scene. Unlike many other settings, conventional camera calibration techniques are not applicable in this case. We present a method that uses certain geometric primitives commonly found in traffic scenes, such as straight and curved lanes, lane markings, and poles in order to recover calibration parameters. We show experimentally that these primitives provide the needed redundancy and are capable of achieving accurate results suitable for most traffic monitoring applications.
|Number of pages
|Transportation Research Part C: Emerging Technologies
|Published - Dec 2007
Bibliographical noteFunding Information:
This work has been supported in part by the National Science Foundation through Grants #IIS-0219863, #CMS-0127893, #CNS-0324864, and #CNS-0224363, the Minnesota Department of Transportation, and the ITS Institute at the University of Minnesota. We would also like to thank the anonymous reviewers for their valuable comments.
- Application systems
- Multi-view reconstruction
- Multi-view stereo
- Traffic monitoring