Assessing Visibility of Individual Transmission Errors in Networked Video

Traditionally, subjective video quality is assessed by user experiments involving quality ratings, pairwise comparisons, or rank ordering, based on the overall impression of quality. Less attention has been paid on assessing the visibility of individual defects. However, many practical applications could benefit from information about subjective visibility of individual packet losses; for example, computational resources could be directed more efficiently to unequal error protection and concealment by focusing in the visually most disturbing artifacts. In this paper, we present a novel subjective methodology for packet loss artifact detection by tapping a touchscreen where a defect is observed. To validate the proposed methodology, the results of a pilot study are presented and analyzed. According to the results, the proposed method can be used to derive qualitatively and statistically meaningful data on the subjective visibility of individual packet loss artifacts.