Back to Search Document details
30th Meeting: Antalya, TR, April 2023 2023-04-25 08:12
AHG4: experiments in preparation of film grain visual tests
Abstract
This document reports about some experiments in preparation of visual tests planned in JVET-AC2022 for FGC SEI visual tests on top of VVC, more specifically the first category which goal is to assess the visual improvement brought by film grain synthesis over a wide range of coding quality. Internal testing in InterDigital labs were performed and indicate that the test planned can show a measurable impact of film grain synthesis on visual quality, which can be translated to bitrate savings.
JVET-AD0276 AHG4: experiments in preparation of film grain visual tests [P. de Lagrange, T. Filoche, F. Urban (InterDigital)] [late]

This document reports about some experiments in preparation of visual tests planned in JVET-AC2022 for FGC SEI visual tests on top of VVC, more specifically the first category which goal is to assess the visual improvement brought by film grain synthesis over a wide range of coding quality. Internal testing in InterDigital labs were performed and indicate that the test planned can show a measurable impact of film grain synthesis on visual quality, which can be translated to bitrate savings.

It is further recommended to add tests at a higher quality range using a direct comparison method, since film grain synthesis at high quality is expected to have a much larger impact on bitrate than for the quality range commonly used for DCR tests.

It is suggested in the contribution to perform two A/B comparison tests, one at transparent point of the coding without FGS, and one at slightly less quality. With regard to the expert viewing in JVET-AD0382, this would mean to identify a “without FGS” coding point which is slightly worse than original, similar to the FGS case.

JVET-AD0276 AHG4: experiments in preparation of film grain visual tests [P. de Lagrange, T. Filoche, F. Urban (InterDigital)] [late]

See notes under 4.5.

Decisions
It is suggested in the contribution to perform two A/B comparison tests, one at transparent point of the coding without FGS, and one at slightly less quality. With regard to the expert viewing in JVET-AD0382, this would mean to identify a “without FGS” coding point which is slightly worse than original, similar to the FGS case.
See notes under 4.5.
Citation