**3. How it influences the subjective experiment?**

The task-based QoE is substantially different from the traditional QoE by different manners. Firstly, as long as a user is able to perform the task we do not have to care if he/she is happy with the overall quality or not up to the level when watching such quality result in fast fatigue of the viewer. Therefore, question about the overall quality does not make much sense. It obviously changes the subjective quality tests significantly.

The second difference between QoE and QoR tests are the source sequences. Let us assume that the task is to recognize if a person on the screen is carrying a gun. In this case more than one source sequence is needed since some alternatives have to be provided. Such alternative sequences have to be very carefully prepared since they should differ from the "gun" sequence by only this one detail. It means that lighting, clouds or any objects at the camera view have to be exactly the same.

The third difference is subjective experiment preparation. In the most traditional QoE experiment a set of parameters of HRC (Hypothetical Reference Circuit) is chosen to produce so called PVS (Processed Video Stream) i.e. a sequence presented to the subjects. A single SRC (Source Reference Circuit) distorted by *n* different HRCs result in generating *n* PVSes. In the QoE all those PVSes are shown to a subject so the impact of HRCs can be analyzed. In case of the QoR such methodology is difficult to use. For example in case of plate recognition if a subject recognizes the plates once, he/she can remember them making the next recognition questionable.

Issues of quality measurements for task-based video are partially addressed in the ITU-T P.912 Recommendation "Subjective video quality assessment methods for recognition tasks" (ITU-T, 2008). This Recommendation introduces basic definitions, methods of testing and ways of conducting psycho-physical experiments (e.g. Multiple Choice Method, Single Answer Method, and Timed Task Method), as well as the distinction between Real-Time- and Viewer-Controlled Viewing scenarios. While these concepts have been introduced specifically for task-based video applications in ITU-T P.912, more research is necessary to validate the methods and refine the data analysis methods. In this chapter we present detailed description for which task-based experiments which methodology can be used.
