Videolab of the Institute for Data Processing

The videolab of the Institute for Data Processing provides a setup for conducting experiments to determine visual and audio-visual quality under standard conditions.

Current research is focused on the development of quality metrics and algorithms modelling the Quality of Experience (QoE) as perceived by humans, in order to avoid time and cost intensive subjective testing for quality estimation.

The videolab participates actively in the European expert network for QoE, Qualinet, and also supports standardization of video codecs by MPEG with subjective testing. Additionally, we offer services based on our experience in visual quality estimation to external, industrial partners.


Equipment

More information about the videolab’s equipment and capabilities.

 


Data Sets & Downloads

Data sets with results from subjective test, supplementary material to our publications and other downloads.


QualityCrowd Framework

QualityCrowd is a framework that allows codec independent quality assessment with a simple web interface, usable with common web browsers.


TUM Multi Format Test Set

48 different video sequences in the formats 576i, 576p, 720p, 1080i and 1080p with diverse content, covering a numerous application areas and provided free of charge for non-commercial use.

 


Publications

Publications based on results of our research activities in the videolab.