Deep convolutional neural models for picture-quality prediction: Challenges and solutions to data-driven image quality assessment

Jongyoo Kim, Hui Zeng, Deepti Ghadiyaram, Sanghoon Lee, Lei Zhang, Alan C. Bovik

Research output: Contribution to journalArticle

60 Citations (Scopus)

Abstract

Convolutional neural networks (CNNs) have been shown to deliver standout performance on a wide variety of visual information processing applications. However, this rapidly developing technology has only recently been applied with systematic energy to the problem of picture-quality prediction, primarily because of limitations imposed by a lack of adequate ground-truth human subjective data. This situation has begun to change with the development of promising data-gathering methods that are driving new approaches to deep-learning-based perceptual picture-quality prediction. Here, we assay progress in this rapidly evolving field, focusing, in particular, on new ways to collect large quantities of ground-truth data and on recent CNN-based picture-quality prediction models that deliver excellent results in a large, real-world, picture-quality database.

Original languageEnglish
Article number8103112
Pages (from-to)130-141
Number of pages12
JournalIEEE Signal Processing Magazine
Volume34
Issue number6
DOIs
Publication statusPublished - 2017 Nov

Fingerprint

Image Quality Assessment
Data-driven
Image quality
Prediction
Neural networks
Neural Networks
Assays
Information Processing
Model
Prediction Model
Energy
Truth

All Science Journal Classification (ASJC) codes

  • Signal Processing
  • Electrical and Electronic Engineering
  • Applied Mathematics

Cite this

Kim, Jongyoo ; Zeng, Hui ; Ghadiyaram, Deepti ; Lee, Sanghoon ; Zhang, Lei ; Bovik, Alan C. / Deep convolutional neural models for picture-quality prediction : Challenges and solutions to data-driven image quality assessment. In: IEEE Signal Processing Magazine. 2017 ; Vol. 34, No. 6. pp. 130-141.
@article{46f5113085d44d03a5931d7791092a4a,
title = "Deep convolutional neural models for picture-quality prediction: Challenges and solutions to data-driven image quality assessment",
abstract = "Convolutional neural networks (CNNs) have been shown to deliver standout performance on a wide variety of visual information processing applications. However, this rapidly developing technology has only recently been applied with systematic energy to the problem of picture-quality prediction, primarily because of limitations imposed by a lack of adequate ground-truth human subjective data. This situation has begun to change with the development of promising data-gathering methods that are driving new approaches to deep-learning-based perceptual picture-quality prediction. Here, we assay progress in this rapidly evolving field, focusing, in particular, on new ways to collect large quantities of ground-truth data and on recent CNN-based picture-quality prediction models that deliver excellent results in a large, real-world, picture-quality database.",
author = "Jongyoo Kim and Hui Zeng and Deepti Ghadiyaram and Sanghoon Lee and Lei Zhang and Bovik, {Alan C.}",
year = "2017",
month = "11",
doi = "10.1109/MSP.2017.2736018",
language = "English",
volume = "34",
pages = "130--141",
journal = "IEEE Signal Processing Magazine",
issn = "1053-5888",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "6",

}

Deep convolutional neural models for picture-quality prediction : Challenges and solutions to data-driven image quality assessment. / Kim, Jongyoo; Zeng, Hui; Ghadiyaram, Deepti; Lee, Sanghoon; Zhang, Lei; Bovik, Alan C.

In: IEEE Signal Processing Magazine, Vol. 34, No. 6, 8103112, 11.2017, p. 130-141.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Deep convolutional neural models for picture-quality prediction

T2 - Challenges and solutions to data-driven image quality assessment

AU - Kim, Jongyoo

AU - Zeng, Hui

AU - Ghadiyaram, Deepti

AU - Lee, Sanghoon

AU - Zhang, Lei

AU - Bovik, Alan C.

PY - 2017/11

Y1 - 2017/11

N2 - Convolutional neural networks (CNNs) have been shown to deliver standout performance on a wide variety of visual information processing applications. However, this rapidly developing technology has only recently been applied with systematic energy to the problem of picture-quality prediction, primarily because of limitations imposed by a lack of adequate ground-truth human subjective data. This situation has begun to change with the development of promising data-gathering methods that are driving new approaches to deep-learning-based perceptual picture-quality prediction. Here, we assay progress in this rapidly evolving field, focusing, in particular, on new ways to collect large quantities of ground-truth data and on recent CNN-based picture-quality prediction models that deliver excellent results in a large, real-world, picture-quality database.

AB - Convolutional neural networks (CNNs) have been shown to deliver standout performance on a wide variety of visual information processing applications. However, this rapidly developing technology has only recently been applied with systematic energy to the problem of picture-quality prediction, primarily because of limitations imposed by a lack of adequate ground-truth human subjective data. This situation has begun to change with the development of promising data-gathering methods that are driving new approaches to deep-learning-based perceptual picture-quality prediction. Here, we assay progress in this rapidly evolving field, focusing, in particular, on new ways to collect large quantities of ground-truth data and on recent CNN-based picture-quality prediction models that deliver excellent results in a large, real-world, picture-quality database.

UR - http://www.scopus.com/inward/record.url?scp=85038027482&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85038027482&partnerID=8YFLogxK

U2 - 10.1109/MSP.2017.2736018

DO - 10.1109/MSP.2017.2736018

M3 - Article

AN - SCOPUS:85038027482

VL - 34

SP - 130

EP - 141

JO - IEEE Signal Processing Magazine

JF - IEEE Signal Processing Magazine

SN - 1053-5888

IS - 6

M1 - 8103112

ER -