Tampere University of Technology

TUTCRIS Research Portal

No-reference visual quality assessment for image inpainting

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Standard

No-reference visual quality assessment for image inpainting. / Voronin, V. V.; Frantc, V. A.; Marchuk, V. I.; Sherstobitov, A. I.; Egiazarian, K.

Image Processing: Algorithms and Systems XIII. SPIE, 2015. 93990U (SPIE Conference Proceedings; Vol. 9399).

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Harvard

Voronin, VV, Frantc, VA, Marchuk, VI, Sherstobitov, AI & Egiazarian, K 2015, No-reference visual quality assessment for image inpainting. in Image Processing: Algorithms and Systems XIII., 93990U, SPIE Conference Proceedings, vol. 9399, SPIE, IS&T/SPIE Electronic Imaging / Image Processing: Algorithms and Systems, 1/01/00. https://doi.org/10.1117/12.2076507

APA

Voronin, V. V., Frantc, V. A., Marchuk, V. I., Sherstobitov, A. I., & Egiazarian, K. (2015). No-reference visual quality assessment for image inpainting. In Image Processing: Algorithms and Systems XIII [93990U] (SPIE Conference Proceedings; Vol. 9399). SPIE. https://doi.org/10.1117/12.2076507

Vancouver

Voronin VV, Frantc VA, Marchuk VI, Sherstobitov AI, Egiazarian K. No-reference visual quality assessment for image inpainting. In Image Processing: Algorithms and Systems XIII. SPIE. 2015. 93990U. (SPIE Conference Proceedings). https://doi.org/10.1117/12.2076507

Author

Voronin, V. V. ; Frantc, V. A. ; Marchuk, V. I. ; Sherstobitov, A. I. ; Egiazarian, K. / No-reference visual quality assessment for image inpainting. Image Processing: Algorithms and Systems XIII. SPIE, 2015. (SPIE Conference Proceedings).

Bibtex - Download

@inproceedings{6f53197b50d848f594cdc4ea4eedc54c,
title = "No-reference visual quality assessment for image inpainting",
abstract = "Inpainting has received a lot of attention in recent years and quality assessment is an important task to evaluate different image reconstruction approaches. In many cases inpainting methods introduce a blur in sharp transitions in image and image contours in the recovery of large areas with missing pixels and often fail to recover curvy boundary edges. Quantitative metrics of inpainting results currently do not exist and researchers use human comparisons to evaluate their methodologies and techniques. Most objective quality assessment methods rely on a reference image, which is often not available in inpainting applications. Usually researchers use subjective quality assessment by human observers. It is difficult and time consuming procedure. This paper focuses on a machine learning approach for no-reference visual quality assessment for image inpainting based on the human visual property. Our method is based on observation that Local Binary Patterns well describe local structural information of the image. We use a support vector regression learned on assessed by human images to predict perceived quality of inpainted images. We demonstrate how our predicted quality value correlates with qualitative opinion in a human observer study. Results are shown on a human-scored dataset for different inpainting methods.",
keywords = "Inpainting, Machine learning, Metric, Quality assessment, SVR, Visual salience",
author = "Voronin, {V. V.} and Frantc, {V. A.} and Marchuk, {V. I.} and Sherstobitov, {A. I.} and K. Egiazarian",
year = "2015",
doi = "10.1117/12.2076507",
language = "English",
isbn = "9781628414899",
series = "SPIE Conference Proceedings",
publisher = "SPIE",
booktitle = "Image Processing: Algorithms and Systems XIII",
address = "United States",

}

RIS (suitable for import to EndNote) - Download

TY - GEN

T1 - No-reference visual quality assessment for image inpainting

AU - Voronin, V. V.

AU - Frantc, V. A.

AU - Marchuk, V. I.

AU - Sherstobitov, A. I.

AU - Egiazarian, K.

PY - 2015

Y1 - 2015

N2 - Inpainting has received a lot of attention in recent years and quality assessment is an important task to evaluate different image reconstruction approaches. In many cases inpainting methods introduce a blur in sharp transitions in image and image contours in the recovery of large areas with missing pixels and often fail to recover curvy boundary edges. Quantitative metrics of inpainting results currently do not exist and researchers use human comparisons to evaluate their methodologies and techniques. Most objective quality assessment methods rely on a reference image, which is often not available in inpainting applications. Usually researchers use subjective quality assessment by human observers. It is difficult and time consuming procedure. This paper focuses on a machine learning approach for no-reference visual quality assessment for image inpainting based on the human visual property. Our method is based on observation that Local Binary Patterns well describe local structural information of the image. We use a support vector regression learned on assessed by human images to predict perceived quality of inpainted images. We demonstrate how our predicted quality value correlates with qualitative opinion in a human observer study. Results are shown on a human-scored dataset for different inpainting methods.

AB - Inpainting has received a lot of attention in recent years and quality assessment is an important task to evaluate different image reconstruction approaches. In many cases inpainting methods introduce a blur in sharp transitions in image and image contours in the recovery of large areas with missing pixels and often fail to recover curvy boundary edges. Quantitative metrics of inpainting results currently do not exist and researchers use human comparisons to evaluate their methodologies and techniques. Most objective quality assessment methods rely on a reference image, which is often not available in inpainting applications. Usually researchers use subjective quality assessment by human observers. It is difficult and time consuming procedure. This paper focuses on a machine learning approach for no-reference visual quality assessment for image inpainting based on the human visual property. Our method is based on observation that Local Binary Patterns well describe local structural information of the image. We use a support vector regression learned on assessed by human images to predict perceived quality of inpainted images. We demonstrate how our predicted quality value correlates with qualitative opinion in a human observer study. Results are shown on a human-scored dataset for different inpainting methods.

KW - Inpainting

KW - Machine learning

KW - Metric

KW - Quality assessment

KW - SVR

KW - Visual salience

U2 - 10.1117/12.2076507

DO - 10.1117/12.2076507

M3 - Conference contribution

SN - 9781628414899

T3 - SPIE Conference Proceedings

BT - Image Processing: Algorithms and Systems XIII

PB - SPIE

ER -