Enhancing VMAF through New Feature Integration and Model Combination

Fan Zhang; Angeliki Katsenou; Christos Bampis; Lukas Krasula; Zhi Li; David Bull

doi:10.1109/PCS50896.2021.9477458

Enhancing VMAF through New Feature Integration and Model Combination

Fan Zhang, Angeliki Katsenou, Christos Bampis, Lukas Krasula, Zhi Li, David Bull

Research output: Chapter in Book/Report/Conference proceeding › Conference Contribution (Conference Proceeding)

7 Citations (Scopus)

116 Downloads (Pure)

Abstract

VMAF is a machine learning based video quality assessment method, originally designed for streaming applications, which combines multiple quality metrics and video features through SVM regression. It offers higher correlation with subjective opinions compared to many conventional quality assessment methods. In this paper we propose enhancements to VMAF through the integration of new video features and alternative quality metrics (selected from a diverse pool) alongside multiple model combination. The proposed combination approach enables training on multiple databases with varying content and distortion characteristics. Our enhanced VMAF method has been evaluated on eight HD video databases, and consistently outperforms the original VMAF model (0.6.1) and other benchmark quality metrics, exhibiting higher correlation with subjective ground truth data.

Original language	English
Title of host publication	2021 Picture Coding Symposium (PCS)
Publisher	Institute of Electrical and Electronics Engineers (IEEE)
Number of pages	5
ISBN (Electronic)	9781665425452
ISBN (Print)	9781665430784
DOIs	https://doi.org/10.1109/PCS50896.2021.9477458
Publication status	Published - 26 Jul 2021

Keywords

eess.IV
cs.CV

Access to Document

10.1109/PCS50896.2021.9477458Licence: Unspecified

Full-text PDF (accepted manuscript)
This is the accepted author manuscript (AAM). The final published version (Version of Record) can be found on the publisher's website. The copyright of any third-party content, such as images, remains with the copyright holder.
Accepted author manuscript, 322 KBLicence: CC BY

Handle.net

Persistent link

Cite this

@inproceedings{eb3b773a1d574aaf967b4269fb129007,

title = "Enhancing VMAF through New Feature Integration and Model Combination",

abstract = " VMAF is a machine learning based video quality assessment method, originally designed for streaming applications, which combines multiple quality metrics and video features through SVM regression. It offers higher correlation with subjective opinions compared to many conventional quality assessment methods. In this paper we propose enhancements to VMAF through the integration of new video features and alternative quality metrics (selected from a diverse pool) alongside multiple model combination. The proposed combination approach enables training on multiple databases with varying content and distortion characteristics. Our enhanced VMAF method has been evaluated on eight HD video databases, and consistently outperforms the original VMAF model (0.6.1) and other benchmark quality metrics, exhibiting higher correlation with subjective ground truth data. ",

keywords = "eess.IV, cs.CV",

author = "Fan Zhang and Angeliki Katsenou and Christos Bampis and Lukas Krasula and Zhi Li and David Bull",

year = "2021",

month = jul,

day = "26",

doi = "10.1109/PCS50896.2021.9477458",

language = "English",

isbn = "9781665430784",

booktitle = "2021 Picture Coding Symposium (PCS)",

publisher = "Institute of Electrical and Electronics Engineers (IEEE)",

address = "United States",

}

TY - GEN

T1 - Enhancing VMAF through New Feature Integration and Model Combination

AU - Zhang, Fan

AU - Katsenou, Angeliki

AU - Bampis, Christos

AU - Krasula, Lukas

AU - Li, Zhi

AU - Bull, David

PY - 2021/7/26

Y1 - 2021/7/26

N2 - VMAF is a machine learning based video quality assessment method, originally designed for streaming applications, which combines multiple quality metrics and video features through SVM regression. It offers higher correlation with subjective opinions compared to many conventional quality assessment methods. In this paper we propose enhancements to VMAF through the integration of new video features and alternative quality metrics (selected from a diverse pool) alongside multiple model combination. The proposed combination approach enables training on multiple databases with varying content and distortion characteristics. Our enhanced VMAF method has been evaluated on eight HD video databases, and consistently outperforms the original VMAF model (0.6.1) and other benchmark quality metrics, exhibiting higher correlation with subjective ground truth data.

AB - VMAF is a machine learning based video quality assessment method, originally designed for streaming applications, which combines multiple quality metrics and video features through SVM regression. It offers higher correlation with subjective opinions compared to many conventional quality assessment methods. In this paper we propose enhancements to VMAF through the integration of new video features and alternative quality metrics (selected from a diverse pool) alongside multiple model combination. The proposed combination approach enables training on multiple databases with varying content and distortion characteristics. Our enhanced VMAF method has been evaluated on eight HD video databases, and consistently outperforms the original VMAF model (0.6.1) and other benchmark quality metrics, exhibiting higher correlation with subjective ground truth data.

KW - eess.IV

KW - cs.CV

U2 - 10.1109/PCS50896.2021.9477458

DO - 10.1109/PCS50896.2021.9477458

M3 - Conference Contribution (Conference Proceeding)

SN - 9781665430784

BT - 2021 Picture Coding Symposium (PCS)

PB - Institute of Electrical and Electronics Engineers (IEEE)

ER -

Enhancing VMAF through New Feature Integration and Model Combination

Abstract

Keywords

Access to Document

Handle.net

Fingerprint

Cite this