This page is provided for historical and archival purposes only. While the seminar dates are correct, we offer no guarantee of informational accuracy or link validity. Contact information for the speakers, hosts and seminar committee are certainly out of date.
The Robotics Institute Carnegie Mellon University
Camera systems with automated zoom lenses are inherently more useful than those with fixed-parameter lenses. Variable-parameter lenses enable us to produce better images by matching the camera's sensing characteristics to the conditions in a scene. They also allow us to make measurements by noting how the scene's image changes as the parameters are varied. The reason variable-parameter lenses are not more commonly used in machine vision is that they are difficult to model for continuous ranges of lens settings.We show in this talk, that traditional modeling approaches cannot capture the complex relationships between control parameters and imaging processes. We then present a methodology for empirically producing accurate camera models for systems with variable-parameter lenses. To demonstrate the effectiveness of our methodology we applied it to produce a variable-parameter, perspective-projection camera model based on Tsai's fixed-parameter camera model. We calibrated and tested our model on two different automated camera systems. In both cases the calibrated model operated across continuous ranges of focus and zoom with an average error of less than 0.14 pixels between the predicted and the measured positions of features in the image plane. We also calibrated and tested our model on one automated camera system across a continuous range of aperture and achieved similar results.
Host: Yangsheng Xu (firstname.lastname@example.org) Appointment: Lalit Katragadda (email@example.com)