Signaling of downsampling filters for chroma from luma intra prediction mode
Methods and apparatuses for performing cross-component intra prediction, including: receiving a coded bitstream; obtaining, from the coded bitstream, a syntax element indicating a downsampling filter used for a cross-component intra prediction mode; obtaining a plurality of reconstructed sample values of a first component which are associated with a pixel of a second component based on the downsampling filter; determining a pixel value of a downsampled pixel of the first component, based on the plurality of reconstructed sample values; determining a pixel value of the pixel of the second component based on the pixel value of the downsampled pixel of the first component; and reconstructing a picture based on the pixel value of the pixel of the second component.
1 . A method for video encoding, the method being performed by at least one processor and comprising:
determining a current chroma block to be coded;
determining that a cross-component intra prediction mode is selected for video coding;
based on determining that the cross-component mode is selected and based on multi-tap downsampling filter being supported, determining pixel values of a plurality of downsampled luma samples for a plurality of chroma samples using a first equation that is associated with the multi-tap downsampling filter being supported in combination with one or more AOMedia Video 1 (“AV1”) filters, wherein the first equation is based on the multi-tap filter being used, pixel values at specific positions, and a rounding value; and
encoding the current chroma block based on at least the plurality of downsampled luma samples.
2 . The method of claim 1 , wherein the multi-tap downsampling filter is one of: (1) a 6-tap filter, (2) a 5-tap filter, (3) a 4-tap filter in which the plurality of luma samples comprise a block co-located with the current chroma block or to one of a left direction, a down direction, and right direction of the current chroma block, (4) a 4-tap filter in which the plurality of luma samples comprise the block co-located with the current chroma block or to one of the right direction, the down direction, or a diagonal direction of the current chroma block; and (5) a 3-tap filter.
3 . The method of claim 1 , wherein when the multi-tap downsampling filter comprises a 6-tap filter, then a pixel value of each downsampled luma sample is determined according to the first equation below:
rec
L
′
(
i
,
j
)
=
[
rec
L
(
2
i
-
1
,
2
j
-
1
)
+
2
·
rec
L
(
2
i
,
2
j
-
1
)
+
rec
L
(
2
i
+
1
,
2
j
-
1
)
+
rec
L
(
2
i
-
1
,
2
j
)
+
2
·
rec
L
(
2
i
,
2
j
)
+
rec
L
(
2
i
+
1
,
2
j
)
+
R
]
≫
3
where the current chroma block is located at a position (i,j), rec′ L indicates the pixel value of the each downsampled luma sample, rec L (x,y) indicates a luma sample located at a luma sample position (x,y), and R indicates the rounding value.
4 . The method of claim 1 , wherein when the multi-tap downsampling filter comprises a 5-tap filter, then the first equation is one of:
rec
L
′
(
i
,
j
)
=
[
rec
L
(
2
i
,
2
j
-
1
)
+
rec
L
(
2
i
-
1
,
2
j
)
+
4
·
rec
L
(
2
i
,
2
j
)
+
rec
L
(
2
i
+
1
,
2
j
)
+
rec
L
(
2
i
,
2
j
+
1
)
+
R
]
≫
3
or
rec
L
′
(
i
,
j
)
=
[
rec
L
(
2
i
,
2
j
-
1
)
+
rec
L
(
2
i
-
1
,
2
j
)
+
12
·
rec
L
(
2
i
,
2
j
)
+
rec
L
(
2
i
+
1
,
2
j
)
+
rec
L
(
2
i
,
2
j
+
1
)
+
R
]
≫
4
where the current chroma block is located at a position (i,j), rec′ L indicates the pixel value of the each downsampled luma sample, rec L (x,y) indicates a luma sample located at a luma sample position (x,y), and R indicates the rounding value.
5 . The method of claim 1 , wherein when the downsampling filter comprises a 4-tap filter, then one luma sample is co-located with the current chroma block.
6 . The method of claim 5 , wherein when the downsampling filter comprises a 4-tap filter, the first equation is one of:
rec
L
′
(
i
,
j
)
=
[
rec
L
(
2
i
-
1
,
2
j
)
+
5
·
rec
L
(
2
i
,
2
j
)
+
rec
L
(
2
i
+
1
,
2
j
)
+
rec
L
(
2
i
,
2
j
+
1
)
+
R
]
≫
3
or
rec
L
′
(
i
,
j
)
=
[
rec
L
(
2
i
-
1
,
2
j
)
+
13
·
rec
L
(
2
i
,
2
j
)
+
rec
L
(
2
i
+
1
,
2
j
)
+
rec
L
(
2
i
,
2
j
+
1
)
+
R
]
≫
4
where the current chroma block is located at a position (i,j), rec′ L indicates the pixel value of the each downsampled luma sample, rec L (x,y) indicates a luma sample located at a luma sample position (x,y), and R indicates the rounding value.
7 . The method of claim 5 , wherein when the downsampling filter comprises a 4-tap filter, the first equation is one of:
rec
L
′
(
i
,
j
)
=
[
rec
L
(
2
i
+
1
,
2
j
)
+
5
·
rec
L
(
2
i
,
2
j
)
+
rec
L
(
2
i
+
1
,
2
j
+
1
)
+
rec
L
(
2
i
,
2
j
+
1
)
+
R
]
≫
3
or
rec
L
′
(
i
,
j
)
=
[
rec
L
(
2
i
+
1
,
2
j
)
+
13
·
rec
L
(
2
i
,
2
j
)
+
rec
L
(
2
i
+
1
,
2
j
+
1
)
+
rec
L
(
2
i
,
2
j
+
1
)
+
R
]
≫
4
where the current chroma block is located at a position (i,j), rec′ L indicates the pixel value of the each downsampled luma sample, rec L (x,y) indicates a luma sample located at a luma sample position (x,y), and R indicates the rounding value.
8 . The method of claim 1 , when the downsampling filter comprises a 3-tap filter, then first equation is one of:
rec
L
′
(
i
,
j
)
=
[
rec
L
(
2
i
-
1
,
2
j
)
+
6
·
rec
L
(
2
i
,
2
j
)
+
rec
L
(
2
i
+
1
,
2
j
)
+
R
]
≫
3
or
rec
L
′
(
i
,
j
)
=
[
rec
L
(
2
i
-
1
,
2
j
)
+
14
·
rec
L
(
2
i
,
2
j
)
+
rec
L
(
2
i
+
1
,
2
j
)
+
R
]
≫
4
where the current chroma block is located at a position (i,j), rec′ L indicates the pixel value of the each downsampled luma sample, rec L (x,y) indicates a luma sample located at a luma sample position (x,y), and R indicates the rounding value.
9 . The method of claim 1 , wherein a syntax element is signaled in at least one from among a sequence header, a sequence parameter set, a picture parameter set, an adaptive parameter set, a video parameter set, a slice header, a picture header, a frame header, a tile header, a coding tree unit header, a superblock header, or a block having a predetermined block size, the syntax element comprising an indication of the multi-tap downsampling filter used in the cross-component intra prediction mode.
10 . A device for performing cross-component intra prediction, the device comprising:
at least one memory configured to store program code; and
at least one processor configured to access the program code and operate as instructed by the program code, the program code including:
obtaining code configured to cause the at least one processor to obtain a current chroma block from a coded bitstream;
determining code configured to cause the at least one processor to determine, from the coded bitstream, that a cross-component intra prediction mode is selected for video decoding;
selecting code configured to cause the at least one processor to determine, based on determining that the cross-component mode is selected and based on multi-tap downsampling filter being supported, pixel values of a plurality of downsampled luma samples for a plurality of chroma samples using a first equation that is associated with the multi-tap downsampling filter being supported in combination with one or more AOMedia Video 1 (′AV1″) filters, wherein the first equation is based on the multi-tap filter being used, pixel values at specific positions, and a rounding value; and
reconstructing code configured to cause the at least one processor to reconstruct the current chroma block based on at least the plurality of downsampled luma samples.
11 . The device of claim 10 , wherein when the multi-tap downsampling filter is one of: (1) a 6-tap filter, (2) a 5-tap filter, (3) a 4-tap filter in which the plurality of luma samples comprise a block co-located with the current chroma block or to one of a left direction, a down direction, and right direction of the current chroma block, (4) a 4-tap filter in which the plurality of luma samples comprise the block co-located with the current chroma block or to one of the right direction, the down direction, or a diagonal direction of the current chroma block; and (5) a 3-tap filter.
12 . The device of claim 10 , wherein when the downsampling filter comprises a 6-tap filter, then a pixel value of each downsampled luma sample is determined according to the first equation below:
rec
L
′
(
i
,
j
)
=
[
rec
L
(
2
i
-
1
,
2
j
-
1
)
+
2
·
rec
L
(
2
i
,
2
j
-
1
)
+
rec
L
(
2
i
+
1
,
2
j
-
1
)
+
rec
L
(
2
i
-
1
,
2
j
)
+
2
·
rec
L
(
2
i
,
2
j
)
+
rec
L
(
2
i
+
1
,
2
j
)
+
R
]
≫
3
where the current chroma block is located at a position (i,j), rec′ L indicates the pixel value of the each downsampled luma sample, rec L (x,y) indicates a luma sample located at a luma sample position (x,y), and R indicates the rounding value.
13 . The device of claim 10 , wherein when the downsampling filter comprises a 5-tap filter, then the first equation is one of:
rec
L
′
(
i
,
j
)
=
[
rec
L
(
2
i
,
2
j
-
1
)
+
rec
L
(
2
i
-
1
,
2
j
)
+
4
·
rec
L
(
2
i
,
2
j
)
+
rec
L
(
2
i
+
1
,
2
j
)
+
rec
L
(
2
i
,
2
j
+
1
)
+
R
]
≫
3
or
rec
L
′
(
i
,
j
)
=
[
rec
L
(
2
i
,
2
j
-
1
)
+
rec
L
(
2
i
-
1
,
2
j
)
+
12
·
rec
L
(
2
i
,
2
j
)
+
rec
L
(
2
i
+
1
,
2
j
)
+
rec
L
(
2
i
,
2
j
+
1
)
+
R
]
≫
4
where the current chroma block is located at a position (i,j), rec′ L indicates the pixel value of the each downsampled luma sample, rec L (x,y) indicates a luma sample located at a luma sample position (x,y), and R indicates the rounding value.
14 . The device of claim 10 , wherein when the downsampling filter comprises a 4-tap filter, then one luma sample is co-located with the current chroma block.
15 . The device of claim 14 , wherein the first equation is one of:
rec
L
′
(
i
,
j
)
=
[
rec
L
(
2
i
-
1
,
2
j
)
+
5
·
rec
L
(
2
i
,
2
j
)
+
rec
L
(
2
i
+
1
,
2
j
)
+
rec
L
(
2
i
,
2
j
+
1
)
+
R
]
≫
3
or
rec
L
′
(
i
,
j
)
=
[
rec
L
(
2
i
-
1
,
2
j
)
+
13
·
rec
L
(
2
i
,
2
j
)
+
rec
L
(
2
i
+
1
,
2
j
)
+
rec
L
(
2
i
,
2
j
+
1
)
+
R
]
≫
4
where the current chroma block is located at a position (i,j), rec′ L indicates the pixel value of the each downsampled luma sample, rec L (x,y) indicates a luma sample located at a luma sample position (x,y), and R indicates the rounding value.
16 . The device of claim 14 , wherein the first equation is one of:
rec
L
′
(
i
,
j
)
=
[
rec
L
(
2
i
+
1
,
2
j
)
+
5
·
rec
L
(
2
i
,
2
j
)
+
rec
L
(
2
i
+
1
,
2
j
+
1
)
+
rec
L
(
2
i
,
2
j
+
1
)
+
R
]
≫
3
or
rec
L
′
(
i
,
j
)
=
[
rec
L
(
2
i
+
1
,
2
j
)
+
13
·
rec
L
(
2
i
,
2
j
)
+
rec
L
(
2
i
+
1
,
2
j
+
1
)
+
rec
L
(
2
i
,
2
j
+
1
)
+
R
]
≫
4
where the current chroma block is located at a position (i,j), rec′ L indicates the pixel value of the each downsampled luma sample, rec L (x,y) indicates a luma sample located at a luma sample position (x,y), and R indicates the rounding value.
17 . The device of claim 10 , wherein when the downsampling filter comprises a 3-tap filter, then the first equation is one of:
rec
L
′
(
i
,
j
)
=
[
rec
L
(
2
i
-
1
,
2
j
)
+
6
·
rec
L
(
2
i
,
2
j
)
+
rec
L
(
2
i
+
1
,
2
j
)
+
R
]
≫
3
or
rec
L
′
(
i
,
j
)
=
[
rec
L
(
2
i
-
1
,
2
j
)
+
14
·
rec
L
(
2
i
,
2
j
)
+
rec
L
(
2
i
+
1
,
2
j
)
+
R
]
≫
4
where the current chroma block is located at a position (i,j), rec′ L indicates the pixel value of the each downsampled luma sample, rec L (x,y) indicates a luma sample located at a luma sample position (x,y), and R indicates the rounding value.
18 . A non-transitory computer-readable medium storing instructions, the instructions comprising: one or more instructions that, when executed by one or more processors of a device for performing cross-component intra prediction, cause the one or more processors to:
perform a conversion between a visual media file and a bitstream of a visual media data according to a format rule,
wherein the bitstream comprises a first syntax element indicating that a cross-component intra prediction mode is selected for video coding; and
wherein the format rule comprises determining pixel values of a plurality of downsampled luma samples for a plurality of chroma samples using a first equation that is associated with the multi-tap downsampling filter being supported based on determining that the cross-component mode is selected and based on multi-tap downsampling filter being supported in combination with one or more AOMedia Video 1 (“AV1”) filters, wherein the first equation is based on the multi-tap filter being used, pixel values at specific positions, and a rounding value.
19 . The non-transitory computer-readable medium of claim 18 , wherein the multi-tap downsampling filter is one of: (1) a 6-tap filter, (2) a 5-tap filter, (3) a 4-tap filter in which the plurality of luma samples comprise a block co-located with the current chroma block or to one of a left direction, a down direction, and right direction of the current chroma block, (4) a 4-tap filter in which the plurality of luma samples comprise the block co-located with the current chroma block or to one of the right direction, the down direction, or a diagonal direction of the current chroma block; and (5) a 3-tap filter.
20 . The non-transitory computer-readable medium of claim 18 , wherein the bitstream comprises the first syntax element in at least one from among a sequence header, a sequence parameter set, a picture parameter set, an adaptive parameter set, a video parameter set, a slice header, a picture header, a frame header, a tile header, a coding tree unit header, a superblock header, or a block having a predetermined block size, and wherein the bitstream further comprises a second syntax element that indicates the multi-tap downsampling filter used in the cross-component intra prediction mode.