Using Many Cameras as One

We illustrate how to consider a network of cameras as a single generalized camera in a framework proposed by Nayar [13]. We derive the discrete structure from motion equations for generalized cameras, and illustrate the corollaries to the epi-polar geometry. This formal mechanism allows one to use a network of cameras as if they were a single imaging device, even when they do not share a common center of projection. Furthermore, an analysis of structure from motion algorithms for this imaging model gives constraints on the optimal design of panoramic imaging systems constructed from multiple cameras.

• Type: Conference
• Year: 2003
• Where: CVPR
• Authors: Robert Pless
• Keywords: Computer Vision, CVPR 2003, Generalized Cameras, Multiple Cameras, Panoramic Imaging Systems, Single Generalized Camera, Single Imaging Device