Academic Press Library in Signal Processing

Chapter 21

Virtual Vision for Camera Networks Research

Faisal Z. Qureshi* and Demetri Terzopoulos^†,    ^*University of Ontario Institute of Technology, Canada,    ^†University of California, Los Angeles, USA

Abstract

Realistic virtual worlds can serve as software laboratories within which vision researchers may prolifically develop and evaluate sophisticated, active machine perception systems. This unorthodox “Virtual Vision” paradigm is posited at the intersection of the fields of computer vision and computer graphics such that virtual reality can subserve computer vision research. An outcome of this paradigm is our work on the rapid development and evaluation of intelligent surveillance systems in large-scale, visually realistic virtual environments ...