Abstract
Image quality assessment (IQA) is an inherent research topic in image processing field for several decades. Recently, machine learning has achieved success in many multimedia tasks and can be applied in IQA. Especially, screen content images (SCIs) is greatly increasing in various applications, but the characteristics of SCIs makes it difficult to directly apply general IQA methods to predict qualities. In this paper, we propose a fast no-reference SCIs quality prediction method. First, we use the convolutional neural networks (CNNs) to predict the quality scores of each patch. Second, we present a SCIs-oriented quality aggregation algorithm for acceleration. Experimental results demonstrate that our method can achieve the high accuracy (0.957) with subjective quality scores, outperforming existing methods. Moreover, our method is computationally appealing, achieving flexible complexity performance by selecting different groups of patches.
| Original language | English |
|---|---|
| Title of host publication | 2018 IEEE International Conference on Multimedia and Expo Workshops, ICMEW 2018 |
| Publisher | Institute of Electrical and Electronics Engineers Inc. |
| ISBN (Electronic) | 9781538641958 |
| DOIs | |
| Publication status | Published - 2018 Nov 28 |
| Event | 2018 IEEE International Conference on Multimedia and Expo Workshops, ICMEW 2018 - San Diego, United States Duration: 2018 Jul 23 → 2018 Jul 27 |
Other
| Other | 2018 IEEE International Conference on Multimedia and Expo Workshops, ICMEW 2018 |
|---|---|
| Country | United States |
| City | San Diego |
| Period | 18/7/23 → 18/7/27 |
Fingerprint
Keywords
- Convolutional Neural Networks
- No-reference Image Quality Assessment
- Screen content images
ASJC Scopus subject areas
- Signal Processing
- Media Technology
Cite this
A fast no-reference screen content image quality prediction using convolutional neural networks. / Cheng, Zhengxue; Takeuchi, Masaru; Kanai, Kenji; Katto, Jiro.
2018 IEEE International Conference on Multimedia and Expo Workshops, ICMEW 2018. Institute of Electrical and Electronics Engineers Inc., 2018. 8551572.Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
}
TY - GEN
T1 - A fast no-reference screen content image quality prediction using convolutional neural networks
AU - Cheng, Zhengxue
AU - Takeuchi, Masaru
AU - Kanai, Kenji
AU - Katto, Jiro
PY - 2018/11/28
Y1 - 2018/11/28
N2 - Image quality assessment (IQA) is an inherent research topic in image processing field for several decades. Recently, machine learning has achieved success in many multimedia tasks and can be applied in IQA. Especially, screen content images (SCIs) is greatly increasing in various applications, but the characteristics of SCIs makes it difficult to directly apply general IQA methods to predict qualities. In this paper, we propose a fast no-reference SCIs quality prediction method. First, we use the convolutional neural networks (CNNs) to predict the quality scores of each patch. Second, we present a SCIs-oriented quality aggregation algorithm for acceleration. Experimental results demonstrate that our method can achieve the high accuracy (0.957) with subjective quality scores, outperforming existing methods. Moreover, our method is computationally appealing, achieving flexible complexity performance by selecting different groups of patches.
AB - Image quality assessment (IQA) is an inherent research topic in image processing field for several decades. Recently, machine learning has achieved success in many multimedia tasks and can be applied in IQA. Especially, screen content images (SCIs) is greatly increasing in various applications, but the characteristics of SCIs makes it difficult to directly apply general IQA methods to predict qualities. In this paper, we propose a fast no-reference SCIs quality prediction method. First, we use the convolutional neural networks (CNNs) to predict the quality scores of each patch. Second, we present a SCIs-oriented quality aggregation algorithm for acceleration. Experimental results demonstrate that our method can achieve the high accuracy (0.957) with subjective quality scores, outperforming existing methods. Moreover, our method is computationally appealing, achieving flexible complexity performance by selecting different groups of patches.
KW - Convolutional Neural Networks
KW - No-reference Image Quality Assessment
KW - Screen content images
UR - http://www.scopus.com/inward/record.url?scp=85059986409&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85059986409&partnerID=8YFLogxK
U2 - 10.1109/ICMEW.2018.8551572
DO - 10.1109/ICMEW.2018.8551572
M3 - Conference contribution
BT - 2018 IEEE International Conference on Multimedia and Expo Workshops, ICMEW 2018
PB - Institute of Electrical and Electronics Engineers Inc.
ER -