Geometric calibration of projector imagery on curved screen based-on subdivision mesh

  • Authors:
  • Jun Zhang;BangPing Wang;XiaoFeng Li

  • Affiliations:
  • Institute of Image & Graphic Department of Computer Science, Sichuan University, Chengdu, China;Institute of Image & Graphic Department of Computer Science, Sichuan University, Chengdu, China;Institute of Image & Graphic Department of Computer Science, Sichuan University, Chengdu, China and Key Laboratory of Fundamental Synthetic Vision Graphics and Image for National Defense, Sichuan ...

  • Venue:
  • GMP'08 Proceedings of the 5th international conference on Advances in geometric modeling and processing
  • Year:
  • 2008

Quantified Score

Hi-index 0.00

Visualization

Abstract

A novel method based on subdivision mesh for geometric calibration problem is proposed, which can correct the projector imagery on smooth curved screen with respect to the screen geometry and the observer's position. Using sparse initial mesh extracted from the image of camera and a special designed subdivision algorithm, we can subdivide the initial mesh into arbitrary precision. Hence, the one-one correspondence of pixels between the frame buffer of projector and image of camera can be established. Using this correspondence, we can transform the projector imagery into any shape through the image of camera. So the geometric calibration can be done easily by mapping all pixels of frame-buffer image to an undistorted mesh on curved screen specified by a laser theodolite. Theoretical analysis and experimental results show that the method can automatically, accurately and rapidly display an undistorted image on curved screen, without any projector and camera registrations or prior knowledge of screen.