Reduced-reference quality assessment for 3D video compression and transmission

3D video services will soon capture the mass market due to the advancement of immersive video technologies (e.g., availability of "3D-Ready" TV monitors at affordable prices) and supportive infrastructure (e.g., high bandwidth communication links). In order to provide demanding customers w...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on consumer electronics 2011-08, Vol.57 (3), p.1185-1193
Main Authors: Hewage, C. T. E. R., Martini, M. G.
Format: Article
Language:English
Subjects:
3D video qualityevaluation
3D video transmission
Channels
Color
Colour
Data mining
Electronics
Image coding
Image color analysis
Image edge detection
Markets
Measurement
PSNR
Receivers
Reduced-Reference quality metric
Studies
Three dimensional
Three dimensional displays
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:3D video services will soon capture the mass market due to the advancement of immersive video technologies (e.g., availability of "3D-Ready" TV monitors at affordable prices) and supportive infrastructure (e.g., high bandwidth communication links). In order to provide demanding customers with a better service over unreliable communication channels, system parameters (e.g., operating conditions of the transmission system) can be updated "on the fly". 3D video quality measured at the receiver side can be used as feedback information to fine tune the system parameters. However, measuring 3D video quality is a challenge due to the complex nature of 3D perceptual attributes such as overall image quality, depth perception, etc. In addition, the use of Full-Reference (FR) quality metrics is not a viable solution due to the need of the original 3D video sequence at the receiver side. This paper therefore proposes a Reduced-Reference (RR) quality metric for colour plus depth 3D video transmission using the extracted edge information of depth maps and extracted information from the corresponding colour image in the areas in the proximity of edges. This work is motivated by the fact that the edges/contours of the depth map can represent different depth levels and identify the main foreground objects of the corresponding colour image and hence can be used in quality evaluations. The performance of the method is evaluated for different compression ratios and network conditions and it shows good results compared to its counterpart FR quality metric, with a lower overhead for side-information.
ISSN:0098-3063
1558-4127
DOI:10.1109/TCE.2011.6018873