Web conferencing and remote desktop applications require the desktop of one computer to be encoded and transmitted to be displayed on a remote screen. However, common video codecs work using the YCbCr color space involving 4:2:0 chrominance sub-sampling, which may lead to annoying artifacts with screen content, e.g. at sharp edges or text.
To automatically detect image areas with a high perceived error, the so-called Perceived Chrominance Sub-sampling Error (PCSE) metric has recently been developed, which shows high correlations to subjective voting.
The task of this research internship is the enhancement of this new metric with regard to automatic classification of image areas exhibiting disturbing chrominance sub-sampling artifacts. The verification of the improved metric should be based on its capability to predict subjective scores for a given data set. Moreover, also visual inspection of image content leading to certain obtained ratings is required.