Signaling of down-sampling information for video bitstreams

1 . A method of processing video data, comprising:

determining, for a conversion between a current video unit of a video and a bitstream of the video, whether information indicating whether down-sampling is performed on the current video unit is included in the bitstream; and

performing the conversion based on the determining,

wherein the information indicates at least one of a method of down-sampling or a re-sampling ratio,

wherein a part of the information is derived directly without signaling, and

wherein the information is included in a supplemental enhancement information (SEI) message associated with the current video unit in the bitstream.

2 . The method of claim 1 , wherein the information in the bitstream excludes at least one of: data regarding whether to perform down-sampling on the current video unit, data regarding the method of down-sampling, or data regarding the re-sampling ratio.

3 . The method of claim 1 , wherein when the information is associated with a sequence, the SEI message comprising the information is before a sequence parameter set (SPS) or a sequence header of the sequence in the bitstream;

when the information is associated with a picture, the SEI message comprising the information is before a picture parameter set (PPS) or a picture header of the picture in the bitstream; and

when the information is associated with a slice, the SEI message comprising the information is before a slice header of the slice in the bitstream.

4 . The method of claim 1 , wherein when the information is associated with a sequence, the SEI message comprising the information is after a sequence parameter set (SPS) or a sequence header of the sequence in the bitstream;

when the information is associated with a picture, the SEI message comprising the information is after a picture parameter set (PPS) or a picture header of the picture in the bitstream; and

when the information is associated with a slice, the SEI message comprising the information is after a slice header of the slice in the bitstream.

5 . The method of claim 1 , wherein the information in the bitstream comprises a flag indicating whether down-sampling is performed on the current video unit.

6 . The method of claim 1 , wherein the information in the bitstream comprises a re-sampling ratio indicating a ratio by which the current video unit is down-sampled, and

wherein down-sampling is performed on the current video unit when the re-sampling ratio is greater than a threshold value.

7 . The method of claim 6 , wherein the re-sampling ratio is determined based on at least one of a width or a height for the current video unit, and

wherein re-sampling ratios for different video unit levels are signalled individually.

8 . The method of claim 1 , wherein the information in the bitstream comprises a syntax element indicating a method of down-sampling performed on the current video unit.

9 . The method of claim 8 , wherein a value of the syntax element equal to 0 indicates that a discrete cosine transform-based interpolation filter is used to perform down-sampling on the current video unit,

the value of the syntax element equal to 1 indicates that a bilinear filter is used to perform down-sampling on the current video unit, and

the value of the syntax element equal to 2 indicates that a bicubic filter is used to perform down-sampling on the current video unit.

10 . The method of claim 8 , wherein a value of the syntax element equal to 3 indicates that a learning-based down-sampling method is used to perform down-sampling on the current video unit, and the value of the syntax element equal to 4 indicates that the method of down-sampling is not specified, and

wherein the learning-based down-sampling method comprises at least one of a convolutional neural network (CNN)-based down-sampling method or a machine learning-based down-sampling method.

11 . The method of claim 1 , wherein the information in the bitstream comprises a flag indicating whether different down-sampling methods are applied for luma components and chroma components of the current video unit.

12 . The method of claim 1 , wherein a first plurality of down-sampling methods for different types of slices are indicated individually in the bitstream, and

wherein a second plurality of down-sampling methods for different temporal layers are indicated individually in the bitstream.

13 . The method of claim 1 , wherein the information in the bitstream indicates a first method of down-sampling performed on the current video unit and a second method of down-sampling performed on the current video unit,

wherein the first method of down-sampling is a learning-based down-sampling method, and the second method of down-sampling uses bilinear filter, bicubic filter, or a discrete cosine transform-based interpolation filter.

14 . The method of claim 1 , wherein the current video unit is a sequence, a picture, a sub-picture, or a slice, and

wherein the information is signaled for the current video unit in a sequence parameter set (SPS), a sequence header, a picture parameter set (PPS), a picture header, a slice header, or an adaptation parameter set (APS).

15 . The method of claim 1 , wherein the information further indicates whether or how a first video unit having a first resolution is determined using a re-sampling process from a second video unit having a second resolution that is different from the first resolution, and

wherein at least one component of the first resolution is lower than at least one component of the second resolution.

16 . The method of claim 1 , wherein the conversion includes encoding the current video unit into the bitstream.

17 . The method of claim 1 , wherein the conversion includes decoding the current video unit from the bitstream.

18 . An apparatus for processing video data comprising a processor and a non-transitory memory with instructions thereon, wherein the instructions upon execution by the processor, cause the processor to:

determine, for a conversion between a current video unit of a video and a bitstream of the video, whether information indicating whether down-sampling is performed on the current video unit is included in the bitstream; and

perform the conversion based on the determining,

wherein the information indicates at least one of a method of down-sampling or a re-sampling ratio,

wherein a part of the information is derived directly without signaling, and

wherein the information is included in a supplemental enhancement information (SEI) message associated with the current video unit in the bitstream.

19 . A non-transitory computer-readable recording medium storing a bitstream of a video which is generated by a method performed by a video processing apparatus, wherein the method comprises:

determining, for a current video unit of the video, whether information indicating whether down-sampling is performed on the current video unit is included in the bitstream; and

generating the bitstream based on the determining,

wherein the information indicates at least one of a method of down-sampling or a re-sampling ratio,

wherein a part of the information is derived directly without signaling, and

wherein the information is included in a supplemental enhancement information (SEI) message associated with the current video unit in the bitstream.

20 . The apparatus of claim 18 , wherein the information in the bitstream comprises a flag indicating whether down-sampling is performed on the current video unit.