	US 12,470,754 B2
	Affine motion model restrictions for memory bandwidth reduction of enhanced interpolation filter
Timofey Mikhailovich Solovyev, Moscow (RU); Maxim Borisovitch Sychev, Moscow (RU); Huanbang Chen, Shenzhen (CN); Alexander Alexandrovich Karabutov, Moscow (RU); Roman Igorevich Chernyak, Moscow (RU); Sergey Yurievich Ikonin, Moscow (RU); Haitao Yang, Shenzhen (CN); and Elena Alexandrovna Alshina, Munich (DE)
Assigned to HUAWEI TECHNOLOGIES CO., LTD., Shenzhen (CN)
Filed by HUAWEI TECHNOLOGIES CO., LTD., Guangdong (CN)
Filed on Feb. 23, 2022, as Appl. No. 17/678,253.
Application 17/678,253 is a continuation of application No. PCT/RU2020/050258, filed on Sep. 30, 2020.
Claims priority of provisional application 62/958,291, filed on Jan. 7, 2020.
Claims priority of application No. PCT/RU2019/000691 (WO), filed on Sep. 30, 2019.
Prior Publication US 2022/0174326 A1, Jun. 2, 2022
Int. Cl. H04N 19/80 (2014.01); H04N 19/105 (2014.01); H04N 19/117 (2014.01); H04N 19/122 (2014.01); H04N 19/132 (2014.01); H04N 19/139 (2014.01); H04N 19/172 (2014.01); H04N 19/176 (2014.01); H04N 19/61 (2014.01); H04N 19/85 (2014.01)

CPC H04N 19/80 (2014.11) [H04N 19/105 (2014.11); H04N 19/117 (2014.11); H04N 19/122 (2014.11); H04N 19/132 (2014.11); H04N 19/139 (2014.11); H04N 19/172 (2014.11); H04N 19/176 (2014.11); H04N 19/61 (2014.11); H04N 19/85 (2014.11)]

18 Claims

1. A method for coding a video implemented in an encoder or a decoder including an enhanced interpolation filter (EIF) for motion compensation, the method comprising:

determining control point motion vectors (CPMVs) for a block according to affine inter-prediction, the block being an affine block or a sub-block of the affine block;

determining a reference area from a reference frame for a sub-block of the block, the sub-block of the block having a predefined sub-block size according to values of the CPMVs, the reference frame separate from the sub-block;

comparing a size of the determined reference area with a predefined threshold;

applying the EIF for motion compensation, comprising deriving a pixel-based motion vector field for the block;

wherein the determined reference area is larger than the predefined threshold, and deriving the pixel-based motion vector field for the block further comprises motion vector clipping,

wherein a motion vector clipping range for the motion vector clipping is determined based on a motion model of the block and a size of the block,

wherein the predefined sub-block size for the EIF is equal to 4×4; wherein dX[0] corresponds to dHorX, dX[1] corresponds to dHorY, dY[0] corresponds to dVerX and dY[1] corresponds to dVerY; wherein determining the reference area for the sub-block with predefined sub-block size according to values of the CPMVs comprises:

setting the value of a variable eifSubblockSize equal to 4;

deriving a variable eifCanBeApplied as follows:

setting the value of eifCanBeApplied to TRUE;

deriving the arrays X[i], Y[i] as follows:

X[0]=0;

X[1]=(eifSubblockSize+1)*(dX[0]+(1<<9));

X[2]=(eifSubblockSize+1)*dY[0];

X[3]=X[1]+X[2];

Y[0]=0;

Y[1]=(eifSubblockSize+1)*dX[1];

Y[2]=(eifSubblockSize+1)*(dY[1]+(1<<9));

Y[3]=Y[1]+Y[2];

setting the value of a variable Xmax equal to maximum of X[i] for i being equal to 0 . . . 3

setting the value of a variable Xmin equal to minimum of X[i] for i being equal to 0 . . . 3

setting the value of a variable Ymax equal to maximum of Y[i] for i being equal to 0 . . . 3

setting the value of a variable Ymin equal to minimum of Y[i] for i being equal to 0 . . . 3

setting the value of a variable W equal to the value of (Xmax−Xmin+(1<<9)−1)>>9

setting the value of a variable H equal to the value of (Ymax−Ymin+(1<<9)−1)>>9.