Hierarchical disparity estimation with energy-based regularization

Hansung Kim, Kwanghoon Sohn

Research output: Contribution to conferencePaper

19 Citations (Scopus)

Abstract

We propose a hierarchical disparity estimation algorithm with energy-based regularization. Initial disparity vectors are obtained from downsampled stereo images using a feature-based region-dividing disparity estimation technique. Dense disparities are estimated from these initial vectors with shape-adaptive windows in full resolution images. Finally, the vector fields are regularized with the minimization of the energy functional which considers both fidelity and smoothness of the fields. The first two steps provide highly reliable disparity vectors, so that local minimum problem can be avoided in regularization step. The proposed algorithm generates accurate disparity map which is smooth inside objects while preserving its discontinuities in boundaries. Experimental results are presented to illustrate the capabilities of the proposed disparity estimation technique.

Original languageEnglish
Pages373-376
Number of pages4
Publication statusPublished - 2003 Dec 16
EventProceedings: 2003 International Conference on Image Processing, ICIP-2003 - Barcelona, Spain
Duration: 2003 Sep 142003 Sep 17

Other

OtherProceedings: 2003 International Conference on Image Processing, ICIP-2003
CountrySpain
CityBarcelona
Period03/9/1403/9/17

Fingerprint

Image resolution

All Science Journal Classification (ASJC) codes

  • Hardware and Architecture
  • Computer Vision and Pattern Recognition
  • Electrical and Electronic Engineering

Cite this

Kim, H., & Sohn, K. (2003). Hierarchical disparity estimation with energy-based regularization. 373-376. Paper presented at Proceedings: 2003 International Conference on Image Processing, ICIP-2003, Barcelona, Spain.
Kim, Hansung ; Sohn, Kwanghoon. / Hierarchical disparity estimation with energy-based regularization. Paper presented at Proceedings: 2003 International Conference on Image Processing, ICIP-2003, Barcelona, Spain.4 p.
@conference{fb663c13ead44a4f93b50d0ed459950c,
title = "Hierarchical disparity estimation with energy-based regularization",
abstract = "We propose a hierarchical disparity estimation algorithm with energy-based regularization. Initial disparity vectors are obtained from downsampled stereo images using a feature-based region-dividing disparity estimation technique. Dense disparities are estimated from these initial vectors with shape-adaptive windows in full resolution images. Finally, the vector fields are regularized with the minimization of the energy functional which considers both fidelity and smoothness of the fields. The first two steps provide highly reliable disparity vectors, so that local minimum problem can be avoided in regularization step. The proposed algorithm generates accurate disparity map which is smooth inside objects while preserving its discontinuities in boundaries. Experimental results are presented to illustrate the capabilities of the proposed disparity estimation technique.",
author = "Hansung Kim and Kwanghoon Sohn",
year = "2003",
month = "12",
day = "16",
language = "English",
pages = "373--376",
note = "Proceedings: 2003 International Conference on Image Processing, ICIP-2003 ; Conference date: 14-09-2003 Through 17-09-2003",

}

Kim, H & Sohn, K 2003, 'Hierarchical disparity estimation with energy-based regularization' Paper presented at Proceedings: 2003 International Conference on Image Processing, ICIP-2003, Barcelona, Spain, 03/9/14 - 03/9/17, pp. 373-376.

Hierarchical disparity estimation with energy-based regularization. / Kim, Hansung; Sohn, Kwanghoon.

2003. 373-376 Paper presented at Proceedings: 2003 International Conference on Image Processing, ICIP-2003, Barcelona, Spain.

Research output: Contribution to conferencePaper

TY - CONF

T1 - Hierarchical disparity estimation with energy-based regularization

AU - Kim, Hansung

AU - Sohn, Kwanghoon

PY - 2003/12/16

Y1 - 2003/12/16

N2 - We propose a hierarchical disparity estimation algorithm with energy-based regularization. Initial disparity vectors are obtained from downsampled stereo images using a feature-based region-dividing disparity estimation technique. Dense disparities are estimated from these initial vectors with shape-adaptive windows in full resolution images. Finally, the vector fields are regularized with the minimization of the energy functional which considers both fidelity and smoothness of the fields. The first two steps provide highly reliable disparity vectors, so that local minimum problem can be avoided in regularization step. The proposed algorithm generates accurate disparity map which is smooth inside objects while preserving its discontinuities in boundaries. Experimental results are presented to illustrate the capabilities of the proposed disparity estimation technique.

AB - We propose a hierarchical disparity estimation algorithm with energy-based regularization. Initial disparity vectors are obtained from downsampled stereo images using a feature-based region-dividing disparity estimation technique. Dense disparities are estimated from these initial vectors with shape-adaptive windows in full resolution images. Finally, the vector fields are regularized with the minimization of the energy functional which considers both fidelity and smoothness of the fields. The first two steps provide highly reliable disparity vectors, so that local minimum problem can be avoided in regularization step. The proposed algorithm generates accurate disparity map which is smooth inside objects while preserving its discontinuities in boundaries. Experimental results are presented to illustrate the capabilities of the proposed disparity estimation technique.

UR - http://www.scopus.com/inward/record.url?scp=0344235347&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0344235347&partnerID=8YFLogxK

M3 - Paper

SP - 373

EP - 376

ER -

Kim H, Sohn K. Hierarchical disparity estimation with energy-based regularization. 2003. Paper presented at Proceedings: 2003 International Conference on Image Processing, ICIP-2003, Barcelona, Spain.