Focus measurement and all in-focus image synthesis for light-field rendering

  • Authors:

  • Keita Takahashi;Akira Kubota;Takeshi Naemura

  • Affiliations:

  • Graduate School of Information Science and Technology, The University of Tokyo, Tokyo, 113-8656 Japan;High Technology Research Center, Kanagawa University, Yokohama, 221-8686 Japan;Graduate School of Information Science and Technology, The University of Tokyo, Tokyo, 113-8656 Japan

  • Venue:
  • Systems and Computers in Japan
  • Year:
  • 2006

Quantified Score

Hi-index 0.00

Visualization

Abstract

Light field rendering is a fundamental method of image-based rendering that synthesizes free-viewpoint images from a set of multiviewpoint images. It is known that this method causes focus-like effects on synthesized images. A problem is that the range of the “focused” depth is too limited in most practical cases. In this paper, we propose a new measure for stably evaluating the “focus” on synthesized images by light-field rendering. We also describe a method of synthesizing an all in-focus image that is clear and sharp throughout the entire scene: First, we synthesize multiple images with different “focused” depths, and then integrate the in-focus parts of those images that can be detected with the “focus measure” we propose. We conducted experiments to demonstrate that our method effectively enhances the quality of the final synthetic images. © 2005 Wiley Periodicals, Inc. Syst Comp Jpn, 37(1): 1–12, 2006; Published online in Wiley InterScience (). DOI 10.1002/scj.20512