Kingston 3D Video Packet Loss Database


Overview

The Kingston 3D Video packet loss database contains 54 3D video sequences (5 to 10 seconds in length) in mobile to HD resolutions. Individual opinion score (OSs) and mean opinion score (MOS) values obtained by the Double Stimulus Continuous Quality Scale (DSCQS) evaluation method are provided.

Nine high quality video sequences which are publicly available (for research purposes) in a range of resolutions were encoded using H.264/AVC video coding standard. H.264 video coding with QP = 30 was used to create the compressed sequences. The transmission of these compressed sequences were simulated over a packet network subjected to random packet losses in order to obtain 54 processed video sequences. The packet loss rates (PLRs) 0%, 1%, 3%, 5% and 10% were considered. The degraded video sequences span the whole perceptual quality range from excellent to bad. The unprocessed reference video sequence was also evaluated since DSCQS method was followed to obtain user`s opinion. 

This database is provided free of charge. 


Goal 

This database was created to perform video quality assessment when 3D video is subjected to random packet losses. The considered packet loss rates are 0%, 1%, 3%, 5% and 10%. There are nine publicly available video sequences to cover a large range of video characteristics. The resolution ranges from mobile to HD.

 More details about this database can be found in following publications.

C.T. Hewage, M.G. Martini, M. B. Brandas, D.V.S.X. De Silva, "A Study on the Perceived Quality of 3-D Video subject to Packet Losses", IEEE International Conference on Communications 2013: IEEE ICC'13 - Workshop on Immersive & Interactive Multimedia Communications over the Future Internet, IEEE Communications Society, Budapest Hungary, pp. 662-666, June 9-13, 2013. 

C.T. Hewage, M.G. Martini, "Quality of Experience for 3D video streaming" in 'IEEE Communications Magazine', vol. 51, no. 5, pp. 101-107, May 2013. 


Possible usage

This 3D video quality database can be used to test a video quality metric precision in an H.264 environment when transmission errors are considered. The high number of video sequences can be used for instance to evaluate the perceived quality depending on the characteristics of the source content.


Publications

Please, cite the following papers in your publications if you use this database in your work:

C.T. Hewage, M.G. Martini, M. B. Brandas, D.V.S.X. De Silva, "A Study on the Perceived Quality of 3-D Video subject to Packet Losses", IEEE International Conference on Communications 2013: IEEE ICC'13 - Workshop on Immersive & Interactive Multimedia Communications over the Future Internet, IEEE Communications Society, Budapest Hungary, pp. 662-666, June 9-13, 2013. 

C.T. Hewage, M.G. Martini, "Quality of Experience for 3D video streaming" in 'IEEE Communications Magazine', vol. 51, no. 5, pp. 101-107, May 2013. 


Contact


Website: WMN Research Group, Kingston University-London

ku43532 (at) kingston (dot) ac (dot) uk or ku3ddatabase@gmail.com