Video Quality Metric Compatible with PSNR Considering Recent Knowledge of Peripheral Characteristics of Human Vision

Telecommunication networks are designed to maintain the high quality of daily media content streaming, and modern video display systems have an increasingly wider field of view. Nowadays, numerous video quality assessment metrics are used in video transmission. However, very few metrics are suitable for wide-angle video transmission, namely the metrics considering how the human visual system perceives temporal frequency in the central and peripheral areas. The presented work analyses existing data on human perception of temporal frequencies in the central area of vision and the peripheral vision. A new method for measuring quality is presented, built on new models of human contrast sensitivity, including thresholds for the visibility of spatio-temporal sinusoidal variations with different spatial dimensions and temporal modulation rates on modern displays. Improvements in the performance of quality metrics are demonstrated when recent knowledge about the functioning of the human visual system is considered. Testing was done on two independent video data sets to demonstrate the algorithm's versatility.