Nonmetric calibration of wide-angle lenses and polycameras.pdf

Non-Metric Calibration of Wide-Angle Lenses and Polycameras ? Rahul Swaminathan and Shree K. Nayar Department of Computer Science, Columbia University New York, New York 10027 Email: srahul@cs.columbia.edu, nayar@cs.columbia.edu Abstract Images taken with wide-angle cameras tend to have severe distortions which pull points towards the optical center. This paper proposes a method for recovering the distor- tion parameters without the use of any calibration objects. The distortions cause straight lines in the scene to appear as curves in the image. Our algorithm seeks to find the distortion parameters that would map the image curves to straight lines. The user selects a small set of points along the image curves. Recovery of the parameters is formu- lated as the minimization of an objective function which is designed to explicitly account for noise in the selected image points. Experimental results are presented for syn- thetic data with different noise levels as well as for real im- ages. Once calibrated, the image streams from these cam- eras can be undistorted in real time using look up tables. We also present an application of this calibration method for wide-angle camera clusters, which we call polycam- eras. We apply our distortion correction technique to a polycamera with four wide-angle cameras to create a high resolution 360 degree panorama in real-time. 1 Introduction In many vision applications, such as surveillance, it is de- sirable to capture the entire region of interest with as few cameras as possible. Wide-angle cameras help in this re- gard, but at the cost of severe image distortions. Wide- angle lenses that adhere to perspective projection would necessitate the use of prohibitively large image detectors. To work around this problem, wide-angle lenses are de- signed to severely bend rays of light around the periphery of the field of view 1 , thus permitting the use of a small image detector (say, a CCD). The effects of the resulting image distortion