VQEG Logo

Video Quality Experts Group (VQEG)

Multimedia Phase I

Validation Test Goals

Multimedia Phase I validated objective video quality models that predicted the quality of VGA, CIF, and QCIF. Model performance was analyzed against both coding-only impairments and transmission errors. Models were submitted in 2007 and VQEG's final report was approved September 12, 2008. The ITU decided that the accuracy of some FR and RR models was sufficient to justify standardization. The ITU decided not pursue an No Reference (NR) model recommendation because the performance of NR models was too low. VQEG will further evaluate NR models in future projects to monitor the progress on this approach. The PSNR implementation that was used as a minimal performance metric was also standardized.

The Multimedia Phase I subjective and objective data have not yet been released.  The video sequences cannot be redistributed both due licensing restrictions on the source material and also due to a multiple party non-disclosure agreement signed by all proponents and ILG.

  • Multimedia Phase I Final Report (pdf)

  • Multimedia Phase I ILG Data Analysis (xls)

  • Multimedia Phase I Testplan (pdf)

  • See the VQEG project archives for other files

Resulting ITU Recommendations

  • ITU-T Rec. J.247 (2008) "Objective perceptual multimedia video quality measurement in the presence of a full reference"

  • ITU-T Rec. J.246 (2008) "Perceptual visual quality measurement techniques for multimedia services over digital cable television networks in the presence of a reduced bandwidth reference"

  • ITU-R Rec. BT.1866, (2010) "Objective perceptual visual quality measurement techniques for broadcasting applications using low definition television in the presence of a full reference signal"

  • ITU-R Rec. BT.1867, (2010) "Objective perceptual visual quality measurement techniques for broadcasting applications using low definition television in the presence of a reduced bandwidth reference"

  • ITU-T Rec J.340 (2010) "Reference algorithm for computing peak signal to noise ratio (PSNR) of a processed video sequence with constant spatial shifts and a constant delay"