Organisation internationale de normalisation

Characteristics of test sequences of screen content

Sequence	Resolution	Frames	Frame rate fps	Chrome format	Bit-depth Per Sample
Doc	1280x800	400	10	4:2:0	8
Gaming	800x600	500	50	4:2:0	8
Slide	1280x800	500	20	4:2:0	8
Web	1280x800	500	25	4:2:0	8

Scene content description of screen content test sequences

Sequence	Brief Description
Doc	A word document is being edited.
Gaming	A flash game playing with moving objects.
Slide	A slide show with all kinds of animations.
Web	Web browsing activities.

A participant remarked that current typical screen content characteristics may use higher resolution than these.

A participant remarked that higher frame rates may be desirable.

A participant requested for 4:4:4 versions to also be made available (which the contributor said they actually already had and could easily provide it within a few days).

TV content with graphics overlays was also suggested to be a valuable type of material to collect for the "screen content" category.

The suggestion was made to create a "Class F" from this content, but Class F would be optional for experiment reporting and would not ordinarily be included in common conditions reported averages for the usual cases (I, RA, and LD for HE and LC).

This suggestion was supported by the group.

It was remarked that it is necessary to ensure that there are no third-party content rights to the content that was captured in the scenes. Any sequences for which some problem is identified in that regard must not included in the Class F testing set.

BoG activity coordinated by O. Au, J. Xu, and H. Yu was requested to further discuss and propose mandates for AHG activity. (See BoG report JCTVC-D458.)

Further work was encouraged (in AHG) to improve the content available for this type of testing.