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 - The International Society for Optical Engineering
PublisherSPIE
Volume9011
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

Other

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

Fingerprint

Depth Map
Noise Reduction
Noise abatement
noise reduction
Enhancement
Real-time
augmentation
commodities
planar structures
Parallel processing systems
Image resolution
Parallel Computing
High Resolution
high resolution

Keywords

  • depth map
  • GPUs
  • high frame rate
  • upsampling

ASJC Scopus subject areas

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

Cite this

Matsumoto, K., Song, C., De Sorbier, F., & Saito, H. (2014). Joint upsampling and noise reduction for real-time depth map enhancement. In Proceedings of SPIE - The International Society for Optical Engineering (Vol. 9011). [901120] SPIE. https://doi.org/10.1117/12.2039190

Joint upsampling and noise reduction for real-time depth map enhancement. / Matsumoto, Kazuki; Song, Chiyoung; De Sorbier, Francois; Saito, Hideo.

Proceedings of SPIE - The International Society for Optical Engineering. Vol. 9011 SPIE, 2014. 901120.

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

Matsumoto, K, Song, C, De Sorbier, F & Saito, H 2014, Joint upsampling and noise reduction for real-time depth map enhancement. in Proceedings of SPIE - The International Society for Optical Engineering. vol. 9011, 901120, SPIE, 25th Stereoscopic Displays and Applications Conference, SD and A 2014, San Francisco, CA, United States, 14/2/3. https://doi.org/10.1117/12.2039190
Matsumoto K, Song C, De Sorbier F, Saito H. Joint upsampling and noise reduction for real-time depth map enhancement. In Proceedings of SPIE - The International Society for Optical Engineering. Vol. 9011. SPIE. 2014. 901120 https://doi.org/10.1117/12.2039190
Matsumoto, Kazuki ; Song, Chiyoung ; De Sorbier, Francois ; Saito, Hideo. / Joint upsampling and noise reduction for real-time depth map enhancement. Proceedings of SPIE - The International Society for Optical Engineering. Vol. 9011 SPIE, 2014.
@inproceedings{ab0400f1495c4c8da8df0a318c60536a,
title = "Joint upsampling and noise reduction for real-time depth map enhancement",
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.",
keywords = "depth map, GPUs, high frame rate, upsampling",
author = "Kazuki Matsumoto and Chiyoung Song and {De Sorbier}, Francois and Hideo Saito",
year = "2014",
doi = "10.1117/12.2039190",
language = "English",
isbn = "9780819499288",
volume = "9011",
booktitle = "Proceedings of SPIE - The International Society for Optical Engineering",
publisher = "SPIE",

}

TY - GEN

T1 - Joint upsampling and noise reduction for real-time depth map enhancement

AU - Matsumoto, Kazuki

AU - Song, Chiyoung

AU - De Sorbier, Francois

AU - Saito, Hideo

PY - 2014

Y1 - 2014

N2 - 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.

AB - 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.

KW - depth map

KW - GPUs

KW - high frame rate

KW - upsampling

UR - http://www.scopus.com/inward/record.url?scp=84901036116&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84901036116&partnerID=8YFLogxK

U2 - 10.1117/12.2039190

DO - 10.1117/12.2039190

M3 - Conference contribution

AN - SCOPUS:84901036116

SN - 9780819499288

VL - 9011

BT - Proceedings of SPIE - The International Society for Optical Engineering

PB - SPIE

ER -