Context Modeling based Depth Image Compression for Distributed Virtual Environment

  • Authors:
  • Paul Bao;Douglas Gourlay;Youfu Li

  • Affiliations:
  • -;-;-

  • Venue:
  • CW '03 Proceedings of the 2003 International Conference on Cyberworlds
  • Year:
  • 2003

Quantified Score

Hi-index 0.00

Visualization

Abstract

Depth images, comprising pixel intensities and depthmap, are viewed as a compact model of 3D scenes andused in 3D image warping for the distributed orcollaborative virtual environment to enable the distributedrendering of complex 3D scenes at relatively low cost. Themajor overhead of the model is the transmission of thedepth images of the initial reference and the subsequentnew views that may be up to a few Mbs in size dependingon the screen resolution. This paper presents efficientcompression techniques specifically designed for the depthimages. Additionally we experiment with the warpedimage quality by reducing the resolution of the depth map.We show that significant compression can be achieved ata cost of a very modest impairment in the perceptualquality of the warped image.