VTT Audiovisual Dataset

Author: VTT Technical Research Center of Finland

Partner: Yes

Contact: Toni Maki (toni.maki@vtt.fi) Martín Varela (martin.varela@vtt.fi)




Subjective scores: true

Total: 125

Ratings: 24

Resolution: 1920x1080, 1280x720, 854x480

Method: DCR


Subjective ratings from 24 participants from Audiovisual assessment campaign. Degraded videos by request (intact original videos are from The Consumer Digital Video Library). In the campaign Audiovisual vote, audio vote and video vote about quality was asked about 125 streamed video samples that were degraded by introducing packet losses. The database can be used to train and validate parametric models with independent variables resolution, amount of movement, packet loss and burstiness.


Go to https://gitlab.willab.fi/qualinet/audiovisualdataset. Then select Download zip or visit the files from Files tab.

Broken link!

The original link was: https://gitlab.willab.fi/qualinet/audiovisualdataset

References and Citation

Please, cite the following paper, if you use this database: [MKD13]


  • MKD13: T. Maki, D. Kukolj, D. Dordevic, and M. Varela, “A reduced-reference parametric model for audiovisual quality of IPTV services,” in 2013 Fifth International Workshop on Quality of Multimedia Experience (QoMEX), 2013, pp. 6–11.