Horizontal Resolution Enhancement of Autostereoscopy Three-Dimensional Displayed Image by Chroma Subpixel Downsampling

Avishai M. Marson, Adrian Stern

Research output: Contribution to journalArticlepeer-review

7 Scopus citations

Abstract

Autostereoscopic displays based on lenticular arrays have reduced resolution in the horizontal direction with respect to the vertical due to the distribution of the horizontal pixels between the different views. One known method is to use a slanted lenticular array in order to increase horizontal resolution at the expense of the vertical. In this paper, we propose a different approach to circumvent the displayed resolution limitation by taking advantage of the lower acuity of the eye to chroma than luma. This is done by downsampling the interlaced multiview display image at the color subpixel level. The chromatic information is in effect subsampled in the horizontal dimension (assuming a vertically arranged subpixel structure of the display), while enhancement of luma resolution is achieved. By appropriate slanting of the lenticular so as to minimize Moiré effects, the chroma is on average sampled evenly over the different views; therefore, chromatic distortion is also minimized. The vertical sampling rate is not decreased in this method; therefore, the total perceived light field resolution is enhanced. We demonstrate an example with an increase in number of views by 40% for a given lenticular and display setup.

Original languageEnglish
Article number6987227
Pages (from-to)800-806
Number of pages7
JournalIEEE/OSA Journal of Display Technology
Volume11
Issue number10
DOIs
StatePublished - 1 Oct 2015

Keywords

  • 3D displays
  • Autostereoscopic
  • Moiré
  • down-sampling
  • lenticular
  • stereoscopic

ASJC Scopus subject areas

  • Electronic, Optical and Magnetic Materials
  • Condensed Matter Physics
  • Electrical and Electronic Engineering

Fingerprint

Dive into the research topics of 'Horizontal Resolution Enhancement of Autostereoscopy Three-Dimensional Displayed Image by Chroma Subpixel Downsampling'. Together they form a unique fingerprint.

Cite this