US 12,355,999 B2
Method and apparatus for efficiently coding residual blocks
Dong Gyu Sim, Seoul (KR); Jong Seok Lee, Seoul (KR); Sea Nae Park, Seoul (KR); Seung Wook Park, Gyeonggi-do (KR); and Wha Pyeong Lim, Gyeonggi-do (KR)
Assigned to Hyundai Motor Company, Seoul (KR); Kia Corporation, Seoul (KR); and Kwangwoon University Industry-Academic Collaboration Foundation, Seoul (KR)
Filed by Hyundai Motor Company, Seoul (KR); Kia Corporation, Seoul (KR); and Kwangwoon University Industry-Academic Collaboration Foundation, Seoul (KR)
Filed on Mar. 28, 2024, as Appl. No. 18/620,697.
Application 18/620,697 is a continuation of application No. 17/438,305, granted, now 11,973,966, previously published as PCT/KR2020/003455, filed on Mar. 12, 2020.
Claims priority of application No. 10-2019-0028364 (KR), filed on Mar. 12, 2019.
Prior Publication US 2024/0244247 A1, Jul. 18, 2024
This patent is subject to a terminal disclaimer.
Int. Cl. H04N 19/46 (2014.01); H04N 19/105 (2014.01); H04N 19/176 (2014.01); H04N 19/186 (2014.01); H04N 19/70 (2014.01); H04N 19/82 (2014.01)
CPC H04N 19/46 (2014.11) [H04N 19/105 (2014.11); H04N 19/176 (2014.11); H04N 19/186 (2014.11); H04N 19/70 (2014.11); H04N 19/82 (2014.11)] 6 Claims
OG exemplary drawing
 
1. A video decoding apparatus comprising at least one processor and a memory, the at least one processor configured to:
decode, from a bitstream, a flag indicating whether residual signals corresponding only to a partial region of a current block have been encoded;
reconstruct transform coefficients by decoding, from the bitstream, transform coefficient information for one subblock of two subblocks split from the current block, when the flag indicates that the residual signals corresponding only to the partial region of the current block have been encoded;
predict the current block to generate a prediction block;
generate a residual block for the current block by performing an inverse quantization and an inverse transform on the transform coefficients in the one subblock for which the transform coefficient information has been decoded;
reconstruct the current block by adding up the prediction block and the residual block; and
set a grid of N samples on the reconstructed current block at regular intervals in horizontal and vertical directions and performing deblock filtering on a boundary between the two subblocks in the current block that coincides with a boundary of the grid;
wherein reconstructing the transform coefficients comprises:
decoding, from the bitstream, one or more syntax elements indicating a partition type of the current block;
determining a partition type for the current block to be split into the two subblocks depending on the one or more syntax elements indicating the partition type of the current block and a width and a height of the current block; and
decoding transform coefficient information corresponding only to the one subblock of the two subblocks that is determined according to the partition type;
wherein the partition type specifies a split ratio and a split direction; and
wherein available split ratios and available split directions for the current block are limited according to the width and height of the current block to provide 4 pixels or more for a height or a width of the one subblock to undergo the inverse transform.