Depth interpolation using tangent planes on superpixels of a color image

Kiyoshi Matsuo, Yoshimitsu Aoki

Research output: Contribution to journalArticle

Abstract

This paper describes a method for accurately interpolating a low-resolution depth image using a high-resolution color image. In our method, first, tangent planes on each superpixel are estimated from the sparse depth information and dense color information. Then, the neighboring superpixels that have smooth-connectable tangent planes are connected, and the image segmentation to smooth surfaces are achieved. Finally, the low-resolution depth image is interpolated using this smooth surface segmentation. In experiments with images from the Middlebury stereo datasets, our method interpolates each image at a high rate, and achieves the lowest error when compared to existing techniques.

Original languageEnglish
Pages (from-to)587-598
Number of pages12
JournalIEEJ Transactions on Electronics, Information and Systems
Volume134
Issue number4
DOIs
Publication statusPublished - 2014

Fingerprint

Interpolation
Color
Image segmentation
Experiments

Keywords

  • Depth image interpolation
  • Image segmentation
  • Superpixel

ASJC Scopus subject areas

  • Electrical and Electronic Engineering

Cite this

Depth interpolation using tangent planes on superpixels of a color image. / Matsuo, Kiyoshi; Aoki, Yoshimitsu.

In: IEEJ Transactions on Electronics, Information and Systems, Vol. 134, No. 4, 2014, p. 587-598.

Research output: Contribution to journalArticle

@article{ca1e122e35574b8091e2485ef5205cfa,
title = "Depth interpolation using tangent planes on superpixels of a color image",
abstract = "This paper describes a method for accurately interpolating a low-resolution depth image using a high-resolution color image. In our method, first, tangent planes on each superpixel are estimated from the sparse depth information and dense color information. Then, the neighboring superpixels that have smooth-connectable tangent planes are connected, and the image segmentation to smooth surfaces are achieved. Finally, the low-resolution depth image is interpolated using this smooth surface segmentation. In experiments with images from the Middlebury stereo datasets, our method interpolates each image at a high rate, and achieves the lowest error when compared to existing techniques.",
keywords = "Depth image interpolation, Image segmentation, Superpixel",
author = "Kiyoshi Matsuo and Yoshimitsu Aoki",
year = "2014",
doi = "10.1541/ieejeiss.134.587",
language = "English",
volume = "134",
pages = "587--598",
journal = "IEEJ Transactions on Electronics, Information and Systems",
issn = "0385-4221",
publisher = "The Institute of Electrical Engineers of Japan",
number = "4",

}

TY - JOUR

T1 - Depth interpolation using tangent planes on superpixels of a color image

AU - Matsuo, Kiyoshi

AU - Aoki, Yoshimitsu

PY - 2014

Y1 - 2014

N2 - This paper describes a method for accurately interpolating a low-resolution depth image using a high-resolution color image. In our method, first, tangent planes on each superpixel are estimated from the sparse depth information and dense color information. Then, the neighboring superpixels that have smooth-connectable tangent planes are connected, and the image segmentation to smooth surfaces are achieved. Finally, the low-resolution depth image is interpolated using this smooth surface segmentation. In experiments with images from the Middlebury stereo datasets, our method interpolates each image at a high rate, and achieves the lowest error when compared to existing techniques.

AB - This paper describes a method for accurately interpolating a low-resolution depth image using a high-resolution color image. In our method, first, tangent planes on each superpixel are estimated from the sparse depth information and dense color information. Then, the neighboring superpixels that have smooth-connectable tangent planes are connected, and the image segmentation to smooth surfaces are achieved. Finally, the low-resolution depth image is interpolated using this smooth surface segmentation. In experiments with images from the Middlebury stereo datasets, our method interpolates each image at a high rate, and achieves the lowest error when compared to existing techniques.

KW - Depth image interpolation

KW - Image segmentation

KW - Superpixel

UR - http://www.scopus.com/inward/record.url?scp=84897997719&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84897997719&partnerID=8YFLogxK

U2 - 10.1541/ieejeiss.134.587

DO - 10.1541/ieejeiss.134.587

M3 - Article

AN - SCOPUS:84897997719

VL - 134

SP - 587

EP - 598

JO - IEEJ Transactions on Electronics, Information and Systems

JF - IEEJ Transactions on Electronics, Information and Systems

SN - 0385-4221

IS - 4

ER -