Edge-preserving directional regularization technique for disparity estimation of stereoscopic images

Mi Hyun Kim, Kwang Hoon Sohn

Research output: Contribution to journalConference article

Abstract

This paper presents a new disparity estimation method using an edge-preserving directional regularization technique. This method smooths disparity vectors in smooth regions and preserves edges in object boundaries without oversmoothing problem. Conclusively, we obtain very close matches between a left image and a right image as well as improved coding efficiency.

Original languageEnglish
Pages (from-to)376-377
Number of pages2
JournalDigest of Technical Papers - IEEE International Conference on Consumer Electronics
Publication statusPublished - 1999 Dec 1

Fingerprint

Image coding

All Science Journal Classification (ASJC) codes

  • Electrical and Electronic Engineering
  • Industrial and Manufacturing Engineering

Cite this

@article{4d049852606f458282e8d7748ec05f6e,
title = "Edge-preserving directional regularization technique for disparity estimation of stereoscopic images",
abstract = "This paper presents a new disparity estimation method using an edge-preserving directional regularization technique. This method smooths disparity vectors in smooth regions and preserves edges in object boundaries without oversmoothing problem. Conclusively, we obtain very close matches between a left image and a right image as well as improved coding efficiency.",
author = "Kim, {Mi Hyun} and Sohn, {Kwang Hoon}",
year = "1999",
month = "12",
day = "1",
language = "English",
pages = "376--377",
journal = "Digest of Technical Papers - IEEE International Conference on Consumer Electronics",
issn = "0747-668X",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - JOUR

T1 - Edge-preserving directional regularization technique for disparity estimation of stereoscopic images

AU - Kim, Mi Hyun

AU - Sohn, Kwang Hoon

PY - 1999/12/1

Y1 - 1999/12/1

N2 - This paper presents a new disparity estimation method using an edge-preserving directional regularization technique. This method smooths disparity vectors in smooth regions and preserves edges in object boundaries without oversmoothing problem. Conclusively, we obtain very close matches between a left image and a right image as well as improved coding efficiency.

AB - This paper presents a new disparity estimation method using an edge-preserving directional regularization technique. This method smooths disparity vectors in smooth regions and preserves edges in object boundaries without oversmoothing problem. Conclusively, we obtain very close matches between a left image and a right image as well as improved coding efficiency.

UR - http://www.scopus.com/inward/record.url?scp=0033347843&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0033347843&partnerID=8YFLogxK

M3 - Conference article

SP - 376

EP - 377

JO - Digest of Technical Papers - IEEE International Conference on Consumer Electronics

JF - Digest of Technical Papers - IEEE International Conference on Consumer Electronics

SN - 0747-668X

ER -