Abstract
The paper proposes a No-Reference (NR) quality assessment measurement originally developed for H.264, used for High Efficiency Video Coding (HEVC). In particular, authors present an investigation of NR metrics to objectively estimate the perceptual quality of a set of processed video sequences. The authors take into account typical distortions introduced by the block-based coding approaches like HEVC codec. The underlying processing used for the quality assessment considers the blockiness caused by the boundaries of each coded block and the blurring as a lack of spatial details. The correlation between the NR quality metrics and the well-known and most widely used objective metric, the Video Quality Model (VQM), is performed to validate the quality prediction accuracy based on the provided scores. The Pearson correlation coefficients obtained stand for promising results for different types of videos.
Cite
CITATION STYLE
Leszczuk, M., Janowski, L., Rouis, K., Papir, Z., & Tahar, J. B. H. (2016). Study of No-Reference Video Quality Metrics for HEVC Compression. Journal of Telecommunications and Information Technology, 1(2016), 22–28. https://doi.org/10.26636/jtit.2016.1.700
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.