Constraint on affine model motion vector

1. A method for video decoding in a decoder, comprising:

decoding prediction information of a current block in a current picture from a coded video bitstream, the current picture occupying an entire picture area of each frame of video in the video bitstream, the prediction information being indicative of an affine model, the current block including three or more control points;

constructing a candidate list such that each candidate predictor in the candidate list comprises three or more motion vector predictors corresponding to the three or more control points of the current block,

wherein, in constructing the candidate list, a candidate predictor is determined as invalid and not added to the candidate list in response to a determination that the candidate predictor does not meet a constraint, wherein the constraint comprises applying different thresholds to a horizontal component and a vertical component of at least one of a first ratio R1, or a second ratio R2 of the candidate predictor;

determining a motion vector for each of the three or more control points of the current block based on a selected candidate predictor from the candidate list, the selected candidate predictor including a corresponding motion vector predictor for each of the three or more control points of the current block, the selected candidate predictor being a block separate from the current block;

determining parameters of the affine model based on the determined motion vectors of the three or more control points, the para meters of the affine model being used to transform between the current block and a reference block in a reference picture that has been reconstructed; and

reconstructing at least a sample of the current block according to the affine model,

wherein the first ratio R1=(|MV1x−MV0x|/W, |MV1y−MV0y|/W), and the second ratio R2=(|MV2x−MV0x|/H, |MV2y−MV0y|/H), where:

W and H denote a width and a height of the current block, respectively,

MV0x is a horizontal component of a motion vector of a first control point of two or more control points for the candidate predictor,

MV1x is a horizontal component of a motion vector of a second control point of the two or more control points for the candidate predictor,

MV2x is a horizontal component of a motion vector of a third control point of the two or more control points for the candidate predictor,

MV0y is a vertical component of the motion vector of the first control point of the two or more control points for the candidate predictor,

MV1y is a vertical component of the motion vector of the second control point of the two or more control points for the candidate predictor,

MV2y is a vertical component of the motion vector of the third control point of the two or more control points for the candidate predictor,

|MV1x−MV0x|/W is a horizontal component of the first ratio R1,

|MV1y−MV0y|/W is a vertical component of the first ratio R1,

|MV2x−MV0x|/H is a horizontal component of the second ratio R2, and

|MV2y−MV0y|/H is a vertical component of the second ratio R2, and

wherein the different thresholds include:

a first threshold applied to a maximum value of the horizontal component and the vertical component of the first ratio R1,

a second threshold applied to a maximum value of the horizontal component and the vertical component of the second ratio R2,

a third threshold applied to a minimum value of the horizontal component and the vertical component of the first ratio R1, and

a fourth threshold applied to a minimum value of the horizontal component and the vertical component of the second ratio R2, and

wherein the first threshold is different from the second threshold, and the third threshold is different from the fourth threshold.

2. The method of claim 1 , further comprising at least one of:

applying the constraint based on at least one of a sequence parameter set, a picture parameter set, and a slice header; or

applying the constraint, which is predefined for constructing the candidate list.

3. An apparatus for video decoding, comprising:

processing circuitry configured to:

reconstruct at least a sample of the current block according to the affine model, wherein

the first ratio R1=(|MV1x−MV0x|/W, |MV1y−MV0y|/W), and the second ratio R2=(|MV2x−MV0x|/H, |MV2y−MV0y|/H), where:

W and H denote a width and a height of the current block, respectively,

MV0x is a horizontal component of a motion vector of a first control point of two or more control points for the candidate predictor,

MV1x is a horizontal component of a motion vector of a second control point of the two or more control points for the candidate predictor,

MV2x is a horizontal component of a motion vector of a third control point of the two or more control points for the candidate predictor,

MV0y is a vertical component of the motion vector of the first control point of the two or more control points for the candidate predictor,

MV1y is a vertical component of the motion vector of the second control point of the two or more control points for the candidate predictor,

MV2y is a vertical component of the motion vector of the third control point of the two or more control points for the candidate predictor,

|MV1x−MV0x|/W is a horizontal component of the first ratio R1,

|MV1y−MV0y|/W is a vertical component of the first ratio R1,

|MV2x−MV0x|/H is a horizontal component of the second ratio R2, and

|MV2y−MV0y|/H is a vertical component of the second ratio R2,

wherein the different thresholds include:

a first threshold applied to a maximum value of the horizontal component and the vertical component of the first ratio R1,

a second threshold applied to a maximum value of the horizontal component and the vertical component of the second ratio R2,

a third threshold applied to a minimum value of the horizontal component and the vertical component of the first ratio R1, and

a fourth threshold applied to a minimum value of the horizontal component and the vertical component of the second ratio R2, and

wherein the first threshold is different from the second threshold, and the third threshold is different from the fourth threshold.

4. The apparatus of claim 3 , wherein the processing circuitry is configured to perform one of:

applying the constraint based on at least one of a sequence parameter set, a picture parameter set, and a slice header; or

applying the constraint, which is predefined for constructing the candidate list.

5. A non-transitory computer-readable medium storing instructions which when executed by a computer for video decoding cause the computer to perform a method comprising:

decoding prediction information of a current block in a current picture from a coded video bitstream, the current picture occupying an entire picture area of each frame of video in the video bitstream, the prediction information being indicative of an affine model, the current block including three or more control points;

constructing a candidate list such that each candidate predictor in the candidate list comprises three or more motion vector predictors corresponding to the three or more control points of the current block,

wherein, in constructing the candidate list, a candidate predictor is determined as invalid and not added to the candidate list in response to a determination that the candidate predictor does not meet a constraint, wherein the constraint comprises applying different thresholds to a horizontal component and a vertical component of at least one of a first ratio R1 or a second ratio R2 of the candidate predictor;

determining a motion vector for each of the three or more control points of the current block based on a selected candidate predictor from the candidate list, the selected candidate predictor including a corresponding motion vector predictor for each of the three or more control points of the current block, the selected candidate predictor being a block separate from the current block;

determining parameters of the affine model based on the determined motion vectors of the three or more control points, the parameters of the affine model being used to transform between the current block and a reference block in a reference picture that has been reconstructed; and

reconstructing at least a sample of the current block according to the affine model,

wherein the first ratio R1=(|MV1x−MV0x|/W, |MV1y−MV0y|/W), and the second ratio R2=(|MV2x−MV0x|/H|MV2y−MV0y|/H), where:

W and H denote a width and a height of the current block, respectively,

MV0x is a horizontal component of a motion vector of a first control point of two or more control points for the candidate predictor,

MV1x is a horizontal component of a motion vector of a second control point of the two or more control points for the candidate predictor,

MV2x is a horizontal component of a motion vector of a third control point of the two or more control points for the candidate predictor,

MV0y is a vertical component of the motion vector of the first control point of the two or more control points for the candidate predictor,

MV1y is a vertical component of the motion vector of the second control point of the two or more control points for the candidate predictor,

MV2y is a vertical component of the motion vector of the third control point of the two or more control points for the candidate predictor,

|MV1x−MV0x|/W is a horizontal component of the first ratio R1,

|MV1y−MV0y|/W is a vertical component of the first ratio R1,

|MV2x−MV0x|/H is a horizontal component of the second ratio R2, and

|MV2y−MV0y|/H is a vertical component of the second ratio R2,

wherein the different thresholds include:

a first threshold applied to a maximum value of the horizontal component and the vertical component of the first ratio R1,

a second threshold applied to a maximum value of the horizontal component and the vertical component of the second ratio R2,

a third threshold applied to a minimum value of the horizontal component and the vertical component of the first ratio R1, and

a fourth threshold applied to a minimum value of the horizontal component and the vertical component of the second ratio R2, and

wherein the first threshold is different from the second threshold, and the third threshold is different from the fourth threshold.

6. The method of claim 1 , wherein the current picture has a spatial dimension of 1920×1080 luminance samples.

7. The method of claim 1 , wherein the constraint indicates:

a limit that defines a value range for a horizontal component of a motion vector difference (MVD); and

a limit that defines a value range for a vertical component of the MVD.