Randomized texture flow estimation using visual similarity

Sunghwan Choi, Dongbo Min, Kwanghoon Sohn

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

Abstract

Exploring underlying texture flows defined with orientation and scale is of a great interest on a variety of vision-related tasks. However, existing methods often fail to capture accurate flows due to over-parameterization of texture deformation or employ a costly global optimization which makes the algorithm computationally demanding. In this paper, we address this inverse problem by casting it as a randomized correspondence search along with a locally-adaptive vector field smoothing. When a small example patch is given as a reference, a randomized deformable matching is performed on the very densely quantized label space, enabling an efficient estimation of texture deformation without quality degeneration, e.g., due to quantization artifacts which often appear in the optimization-driven discrete approaches. The visual similarity with respect to the deformation parameters is directly measured with an input texture image on an appearance space. The locally-adaptive smoothing is then applied to the intermediate flow field, resulting in a good continuation of the resultant texture flow. Experimental results on both synthetic and natural images show that the proposed method improves the performance in terms of both runtime efficiency and/or visual quality, compared to the existing methods.

Original languageEnglish
Title of host publication2014 IEEE International Conference on Image Processing, ICIP 2014
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages4662-4666
Number of pages5
ISBN (Electronic)9781479957514
DOIs
Publication statusPublished - 2014 Jan 28

Publication series

Name2014 IEEE International Conference on Image Processing, ICIP 2014

Fingerprint

Textures
Global optimization
Parameterization
Inverse problems
Labels
Flow fields
Casting

All Science Journal Classification (ASJC) codes

  • Computer Vision and Pattern Recognition

Cite this

Choi, S., Min, D., & Sohn, K. (2014). Randomized texture flow estimation using visual similarity. In 2014 IEEE International Conference on Image Processing, ICIP 2014 (pp. 4662-4666). [7025945] (2014 IEEE International Conference on Image Processing, ICIP 2014). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICIP.2014.7025945
Choi, Sunghwan ; Min, Dongbo ; Sohn, Kwanghoon. / Randomized texture flow estimation using visual similarity. 2014 IEEE International Conference on Image Processing, ICIP 2014. Institute of Electrical and Electronics Engineers Inc., 2014. pp. 4662-4666 (2014 IEEE International Conference on Image Processing, ICIP 2014).
@inproceedings{768a4458b9654f399bde3b8a42b75ee8,
title = "Randomized texture flow estimation using visual similarity",
abstract = "Exploring underlying texture flows defined with orientation and scale is of a great interest on a variety of vision-related tasks. However, existing methods often fail to capture accurate flows due to over-parameterization of texture deformation or employ a costly global optimization which makes the algorithm computationally demanding. In this paper, we address this inverse problem by casting it as a randomized correspondence search along with a locally-adaptive vector field smoothing. When a small example patch is given as a reference, a randomized deformable matching is performed on the very densely quantized label space, enabling an efficient estimation of texture deformation without quality degeneration, e.g., due to quantization artifacts which often appear in the optimization-driven discrete approaches. The visual similarity with respect to the deformation parameters is directly measured with an input texture image on an appearance space. The locally-adaptive smoothing is then applied to the intermediate flow field, resulting in a good continuation of the resultant texture flow. Experimental results on both synthetic and natural images show that the proposed method improves the performance in terms of both runtime efficiency and/or visual quality, compared to the existing methods.",
author = "Sunghwan Choi and Dongbo Min and Kwanghoon Sohn",
year = "2014",
month = "1",
day = "28",
doi = "10.1109/ICIP.2014.7025945",
language = "English",
series = "2014 IEEE International Conference on Image Processing, ICIP 2014",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "4662--4666",
booktitle = "2014 IEEE International Conference on Image Processing, ICIP 2014",
address = "United States",

}

Choi, S, Min, D & Sohn, K 2014, Randomized texture flow estimation using visual similarity. in 2014 IEEE International Conference on Image Processing, ICIP 2014., 7025945, 2014 IEEE International Conference on Image Processing, ICIP 2014, Institute of Electrical and Electronics Engineers Inc., pp. 4662-4666. https://doi.org/10.1109/ICIP.2014.7025945

Randomized texture flow estimation using visual similarity. / Choi, Sunghwan; Min, Dongbo; Sohn, Kwanghoon.

2014 IEEE International Conference on Image Processing, ICIP 2014. Institute of Electrical and Electronics Engineers Inc., 2014. p. 4662-4666 7025945 (2014 IEEE International Conference on Image Processing, ICIP 2014).

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

TY - GEN

T1 - Randomized texture flow estimation using visual similarity

AU - Choi, Sunghwan

AU - Min, Dongbo

AU - Sohn, Kwanghoon

PY - 2014/1/28

Y1 - 2014/1/28

N2 - Exploring underlying texture flows defined with orientation and scale is of a great interest on a variety of vision-related tasks. However, existing methods often fail to capture accurate flows due to over-parameterization of texture deformation or employ a costly global optimization which makes the algorithm computationally demanding. In this paper, we address this inverse problem by casting it as a randomized correspondence search along with a locally-adaptive vector field smoothing. When a small example patch is given as a reference, a randomized deformable matching is performed on the very densely quantized label space, enabling an efficient estimation of texture deformation without quality degeneration, e.g., due to quantization artifacts which often appear in the optimization-driven discrete approaches. The visual similarity with respect to the deformation parameters is directly measured with an input texture image on an appearance space. The locally-adaptive smoothing is then applied to the intermediate flow field, resulting in a good continuation of the resultant texture flow. Experimental results on both synthetic and natural images show that the proposed method improves the performance in terms of both runtime efficiency and/or visual quality, compared to the existing methods.

AB - Exploring underlying texture flows defined with orientation and scale is of a great interest on a variety of vision-related tasks. However, existing methods often fail to capture accurate flows due to over-parameterization of texture deformation or employ a costly global optimization which makes the algorithm computationally demanding. In this paper, we address this inverse problem by casting it as a randomized correspondence search along with a locally-adaptive vector field smoothing. When a small example patch is given as a reference, a randomized deformable matching is performed on the very densely quantized label space, enabling an efficient estimation of texture deformation without quality degeneration, e.g., due to quantization artifacts which often appear in the optimization-driven discrete approaches. The visual similarity with respect to the deformation parameters is directly measured with an input texture image on an appearance space. The locally-adaptive smoothing is then applied to the intermediate flow field, resulting in a good continuation of the resultant texture flow. Experimental results on both synthetic and natural images show that the proposed method improves the performance in terms of both runtime efficiency and/or visual quality, compared to the existing methods.

UR - http://www.scopus.com/inward/record.url?scp=84983122740&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84983122740&partnerID=8YFLogxK

U2 - 10.1109/ICIP.2014.7025945

DO - 10.1109/ICIP.2014.7025945

M3 - Conference contribution

T3 - 2014 IEEE International Conference on Image Processing, ICIP 2014

SP - 4662

EP - 4666

BT - 2014 IEEE International Conference on Image Processing, ICIP 2014

PB - Institute of Electrical and Electronics Engineers Inc.

ER -

Choi S, Min D, Sohn K. Randomized texture flow estimation using visual similarity. In 2014 IEEE International Conference on Image Processing, ICIP 2014. Institute of Electrical and Electronics Engineers Inc. 2014. p. 4662-4666. 7025945. (2014 IEEE International Conference on Image Processing, ICIP 2014). https://doi.org/10.1109/ICIP.2014.7025945