Back to Search
Document details
CE1.6.1: Coding large size CUs
Abstract
This contribution reports the results of CE1.6.1.1 and CE1.6.1.2 for handling a coding unit (CU) of width or height larger than the maximum transform size. The proposed methods reportedly remove extra implementation for processing the CUs containing more than one transform unit (TU), which is employed for coding the CUs of width or height larger than the maximum transform size. In CE1.6.1.1, the CU is inferred to be further partitioned by a quadtree split when the width or height of the current CU is larger than the maximum transform size. In CE1.6.1.2, the CU is inferred to have zero residual in a B or P slice when the width or height of the current CU is larger than the maximum transform size. It is reported that each CU always contains one transform unit (TU) and extra implementation to process the CUs containing more than one TU is not needed using the proposed methods in both CE tests. CE1.6.1.1 reportedly leads to -0.01%, 0.90%, and 0.64% luma BD-rates for the AI, RA, and LB settings, respectively, under the VTM-1.0 common test conditions (CTCs) with all BMS tools off. CE1-6.1.2 reportedly leads to -0.01%, 0.07%, and 0.02% luma BD-rates for the AI, RA, and LB settings, respectively, under the VTM-1.0 CTCs with all BMS tools off.
JVET-K0227 CE1.6.1: Coding large size CUs [S.-T. Hsiang, C.-Y. Chen, C.-Y. Lai, Y.-W. Huang, S.-M. Lei (MediaTek)]
References:
PATENTS:
US20250113047A1
0.72
2025-04-03
TWI687092B
0.72
2020-03-01
US9756359B2
0.70
2017-09-05
US12184879B2
0.70
2024-12-31
CN111149359B
0.28
2023-11-17
CN108781293B
0.24
2022-05-27
AU2016203628B2
0.22
2018-05-31
KR20220080062A
0.20
2022-06-14
Decisions
JVET-K0227 CE1.6.1: Coding large size CUs [S.-T. Hsiang, C.-Y. Chen, C.-Y. Lai, Y.-W. Huang, S.-M. Lei (MediaTek)]
Citation