Interpolation filter clipping for sub-picture motion vectors
A video coding mechanism is disclosed. The mechanism includes receiving a bitstream comprising a current picture including a sub-picture coded according to inter-prediction. A motion vector for a block of the sub-picture is determined. A clipping function is applied to sample locations in a reference block to support application of an interpolation filter when the motion vector points outside of the sub-picture and when a flag is set to indicate the sub-picture is treated as a picture. The interpolation filter is applied to results of the clipping function to obtain a predicted sample value. The block is decoded based on the predicted sample value. The block is forwarded for display as part of a decoded video sequence.
1 . A method implemented by a decoder, comprising:
receiving a current picture including a sub-picture with a block coded according to inter-prediction;
determining a motion vector for the block;
applying a clipping function to sample locations in a reference block during application of an interpolation process when a flag is set to indicate the sub-picture is treated as a picture;
applying the interpolation process to results of the clipping function to obtain a predicted sample value; and
decoding the block based on the predicted sample value,
wherein the interpolation process includes a luma sample bilinear interpolation process, wherein the block includes a block of luma samples, wherein the predicted sample value includes a predicted luma sample value, wherein the luma sample bilinear interpolation process receives inputs including a luma location in full sample units (xInt L , yInt L ), and wherein the luma sample bilinear interpolation process outputs the predicted luma sample value (predSampleLX L ), and
wherein the clipping function is applied to the sample locations according to:
xInti=Clip3(SubPicLeftBoundaryPos, SubPicRightBoundaryPos, xInt L +i ), and
yInti=Clip3(SubPicTopBoundaryPos, SubPicBotBoundaryPos, yInt L +i ),
where xInti and yInti are a sample location at index i, SubPicRightBoundaryPos is a position of a right boundary of the sub-picture, SubPicLeftBoundaryPos is a position of a left boundary of the sub-picture, SubPicTopBoundaryPos is a position of a top boundary of the sub-picture, and
where SubPicBotBoundaryPos is a position of a bottom boundary of the sub-picture, and Clip3 is the clipping function according to:
Clip
3
(
x
,
y
,
z
)
=
{
x
;
z
<
x
y
;
z
>
y
z
;
otherwise
where x, y, and z are numerical input values.
2 . The method of claim 1 , wherein the clipping function is applied to the sample locations when subpic_treated_as_pic_flag[SubPicIdx] is equal to one, where subpic_treated_as_pic_flag is the flag set to indicate the sub-picture is treated as a picture, SubPicIdx is an index of the sub-picture.
3 . The method of claim 1 , wherein the interpolation process includes a luma sample interpolation filtering process.
4 . A method implemented by an encoder, comprising:
partitioning a current picture into a sub-picture and the sub-picture into a block;
determining to encode the block according to inter-prediction;
selecting a motion vector to encode the block;
applying a clipping function to sample locations in a reference block to support application of an interpolation process when the motion vector points outside of the sub-picture and when a flag is set to indicate the sub-picture is treated as a picture;
applying the interpolation process to results of the clipping function to obtain a predicted sample value;
encoding the block into a bitstream based on the predicted sample value and the motion vector; and
storing the bitstream for communication toward a decoder,
wherein the interpolation process includes a luma sample bilinear interpolation process, wherein the block includes a block of luma samples, wherein the predicted sample value includes a predicted luma sample value, wherein the luma sample bilinear interpolation process receives inputs including a luma location in full sample units (xInt L , yInt L ), and wherein the luma sample bilinear interpolation process outputs the predicted luma sample value (predSampleLX L ), and
wherein the clipping function is applied to the sample locations according to:
xInti=Clip3(SubPicLeftBoundaryPos, SubPicRightBoundaryPos, xInt L +i ), and
yInti=Clip3(SubPicTopBoundaryPos, SubPicBotBoundaryPos, yInt L +i ),
where xInti and yInti are a sample location at index i, SubPicRightBoundaryPos is a position of a right boundary of the sub-picture, SubPicLeftBoundaryPos is a position of a left boundary of the sub-picture, SubPicTopBoundaryPos is a position of a top boundary of the sub-picture, and
where SubPicBotBoundaryPos is a position of a bottom boundary of the sub-picture, and Clip3 is the clipping function according to:
Clip
3
(
x
,
y
,
z
)
=
{
x
;
z
<
x
y
;
z
>
y
z
;
otherwise
where x, y, and z are numerical input values.
5 . The method of claim 4 , wherein the clipping function is applied to the sample locations when subpic_treated_as_pic_flag[SubPicIdx] is equal to one, where subpic_treated_as_pic_flag is the flag set to indicate the sub-picture is treated as a picture and SubPicIdx is an index of the sub-picture.
6 . The method of claim 4 , wherein the interpolation process includes a luma sample interpolation filtering process.
7 . A decoder, comprising:
a receiver configured to receive a current picture including a sub-picture with a block coded according to inter-prediction; and
one or more processors coupled to the receiver and configured to:
determine a motion vector for the block;
apply a clipping function to sample locations in a reference block during application of an interpolation process when a flag is set to indicate the sub-picture is treated as a picture;
apply the interpolation process to results of the clipping function to obtain a predicted sample value; and
decode the block based on the predicted sample value,
wherein the interpolation process includes a luma sample bilinear interpolation process, wherein the block includes a block of luma samples, wherein the predicted sample value includes a predicted luma sample value, wherein the luma sample bilinear interpolation process receives inputs including a luma location in full sample units (xInt L , yInt L ), and wherein the luma sample bilinear interpolation process outputs the predicted luma sample value (predSampleLX L ), and
wherein the clipping function is applied to the sample locations according to:
xInti=Clip3(SubPicLeftBoundaryPos, SubPicRightBoundaryPos, xInt L +i ), and
yInti=Clip3(SubPicTopBoundaryPos, SubPicBotBoundaryPos, yInt L +i ),
where xInti and yInti are a sample location at index i, SubPicRightBoundaryPos is a position of a right boundary of the sub-picture, SubPicLeftBoundaryPos is a position of a left boundary of the sub-picture, SubPicTopBoundaryPos is a position of a top boundary of the sub-picture, and
where SubPicBotBoundaryPos is a position of a bottom boundary of the sub-picture, and Clip3 is the clipping function according to:
Clip
3
(
x
,
y
,
z
)
=
{
x
;
z
<
x
y
;
z
>
y
z
;
otherwise
where x, y, and z are numerical input values.
8 . The decoder of claim 7 , wherein the clipping function is applied to the sample locations when subpic_treated_as_pic_flag[SubPicIdx] is equal to one, where subpic_treated_as_pic_flag is the flag set to indicate the sub-picture is treated as a picture and SubPicIdx is an index of the sub-picture.
9 . The decoder of claim 7 , wherein the interpolation process includes a luma sample interpolation filtering process.