JCTVC-M0389 Non-SCE3: Crosscheck of JCTVC-M0090 on low-pass filter for combined inter prediction [T.-D. Chuang, Y.-W. Huang (MediaTek)] [late]
JCTVC-M0092 Non SCE3: Simplified design for combined prediction (test 3.2) [E. Alshina, A. Alshin (Samsung)]
Asserted simplifications for combined prediction tested in SCE4 as test 3.2 are proposed in this contribution. With some performance degradation, encoding time was reportedly reduced from 114% to 104%. The worst case memory band-width for combined prediction with suggested modifications is reportedly lower than HEVC in the worst case. The modified version of test 3.2 reportedly provides 1.0% luma and 4.0% chroma BD-rate gain compared to SHM1.0 (IntraBL framework).
To add abstract.
By restricting 8x8 bi-pred, the worst-case memory bandwidth for the enhancement layer is claimed to be within the margin of current SHM. Average gain approx. 1% compared to SHM.
According to JCTVC-M0220/ JCTVC-M0297, disabling bi-pred entirely seems to be a better tradeoff without increasing computations.
Dostları ilə paylaş: |
