Joint upsampling and noise reduction for real-time depth map enhancement

Kazuki Matsumoto, Chiyoung Song, Francois De Sorbier, Hideo Saito

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)

Abstract

An efficient system that upsamples depth map captured by Microsoft Kinect while jointly reducing the effect of noise is presented. The upsampling is carried by detecting and exploiting the piecewise locally planar structures of the downsampled depth map, based on corresponding high-resolution RGB image. The amount of noise is reduced by accumulating the downsampled data simultaneously. By benefiting from massively parallel computing capability of modern commodity GPUs, the system is able to maintain high frame rate. Our system is observed to produce the upsampled depth map that is very close to the original depth map both visually and mathematically.

Original languageEnglish
Title of host publicationProceedings of SPIE-IS and T Electronic Imaging - Stereoscopic Displays and Applications XXV
PublisherSPIE
ISBN (Print)9780819499288
DOIs
Publication statusPublished - 2014
Event25th Stereoscopic Displays and Applications Conference, SD and A 2014 - San Francisco, CA, United States
Duration: 2014 Feb 32014 Feb 5

Publication series

NameProceedings of SPIE - The International Society for Optical Engineering
Volume9011
ISSN (Print)0277-786X
ISSN (Electronic)1996-756X

Other

Other25th Stereoscopic Displays and Applications Conference, SD and A 2014
Country/TerritoryUnited States
CitySan Francisco, CA
Period14/2/314/2/5

Keywords

  • GPUs
  • depth map
  • high frame rate
  • upsampling

ASJC Scopus subject areas

  • Electronic, Optical and Magnetic Materials
  • Condensed Matter Physics
  • Computer Science Applications
  • Applied Mathematics
  • Electrical and Electronic Engineering

Fingerprint

Dive into the research topics of 'Joint upsampling and noise reduction for real-time depth map enhancement'. Together they form a unique fingerprint.

Cite this