Filter shape switching

1 . A method of video decoding, comprising:

receiving a video bitstream corresponding to a plurality of samples of video data, the plurality of samples including a first sample and a second sample;

reconstructing the first sample using a non-linear mapping-based filter with a first filter shape configuration of a plurality of filter shape configurations, wherein:

the plurality of filter shape configurations includes at least two filter shape configurations that are based on a same geometric shape and include a same number of filter taps, and

the filter taps of the at least two filter shape configurations are located at different positions;

selecting a second filter shape configuration of the plurality of filter shape configurations; and

reconstructing the second sample based on the non-linear mapping-based filter with the second filter shape configuration.

2 . The method of claim 1 , wherein the non-linear mapping-based filter includes at least one of a cross-component sample offset (CCSO) filter, and a local sample offset (LSO) filter.

3 . The method of claim 1 , wherein the first filter shape configuration differs from the second filter shape configuration by a distance from filter tap locations to a center of the filter tap locations.

4 . The method of claim 1 , wherein the plurality of filter shape configurations include at least one filter shape configuration of: a cross geometric shape of filter tap locations, and a rectangular geometric shape of filter tap locations.

5 . The method of claim 1 , wherein the first filter shape configuration and the second filter shape configuration are symmetrical.

6 . The method of claim 1 , further comprising:

decoding an index from the video bitstream, the index indicating the second filter shape configuration; and

selecting the second filter shape configuration based on the index.

7 . The method of claim 6 , wherein the index is decoded from at least one of a block level, a video parameter set (VPS), a sequence parameter set (SPS), a picture parameter set (PPS), an adaptation parameter set (APS), a slice header, a tile header, and a frame header.

8 . The method of claim 1 , wherein the second filter shape configuration is selected at a picture level, wherein the first sample is in a first picture of the video bitstream, and wherein the second sample is in a second picture of the video bitstream.

9 . The method of claim 1 , wherein the second filter shape configuration is selected at a block level, wherein the first sample is in a first block in a picture of the video bitstream, and wherein the second sample is in a second block in the picture of the video bitstream.

10 . The method of claim 1 , wherein the reconstructing the first sample in the video bitstream based on the non-linear mapping-based filter with the first filter shape configuration comprises:

performing a pre-processing operation on samples at filter tap locations corresponding to the first filter shape configuration to generate pre-processed samples; and

determining an offset distance to apply on the first sample based on the pre-processed samples.

11 . The method of claim 10 , wherein the performing the pre-processing operation comprises at least one of:

calculating an average sample value at two or more of the filter tap locations as a pre-processed sample; and

applying a filter on a sample at a filter tap location of the filter tap locations to generate a filtered sample as the pre-processed sample.

12 . A method of video encoding, comprising:

receiving video data;

selecting, from a plurality of filter shape configurations, a filter shape configuration of a non-linear mapping-based filter to be applied to a first sample in the video data, wherein:

the plurality of filter shape configurations includes a first filter shape configuration and a second filter shape configuration that are based on a same geometric shape and include a same number of filter taps, and

the filter taps of the plurality of filter shape configurations are located at different positions; and

encoding, in a video bitstream, the first sample and the selected filter shape configuration for the first sample.

13 . The method of claim 12 , wherein the non-linear mapping-based filter includes at least one of a cross-component sample offset (CCSO) filter and a local sample offset (LSO) filter.

14 . The method of claim 12 , wherein the first filter shape configuration differs from the second filter shape configuration by a distance from filter tap locations to a center of the filter tap locations.

15 . The method of claim 12 , wherein the plurality of filter shape configurations includes at least one filter shape configuration of: a cross geometric shape of filter tap locations, and a rectangular geometric shape of filter tap locations.

16 . The method of claim 12 , wherein the first filter shape configuration and the second filter shape configuration are symmetrical.

17 . The method of claim 12 , further comprising encoding, in the video bitstream, an index indicative of the selected filter shape configuration for the first sample.

18 . The method of claim 17 , wherein the index is signaled at a picture level in the video bitstream, and wherein the first sample is in a picture of the video data.

19 . The method of claim 17 , wherein the index is signaled at a block level of the video bitstream, and wherein the first sample is in a block in a picture of the video data.

20 . A non-transitory computer-readable storage medium storing one or more instructions and a video bitstream that is generated by a video encoding method, the one or more instructions when executed by a processor, cause a computing system to perform the video encoding method, the encoding method comprising:

receiving video data;

selecting, from a plurality of filter shape configurations, a filter shape configuration of a non-linear mapping-based filter to be applied to a first sample in the video data, wherein:

the plurality of filter shape configurations includes a first filter shape configuration and a second filter shape configuration that are based on a same geometric shape and include a same number of filter taps, and

the filter taps of the plurality of filter shape configurations are located at different positions; and

encoding, in the video bitstream, the first sample and the selected filter shape configuration for the first sample.