Video or image coding on basis of conditionally parsed ALF model and reshaping model
According to the disclosure of the present document, reshaper data and/or an ALF parameter can be signaled adaptively from a parameter set, thereby reducing the amount of data to be signaled for video/image coding and improving coding efficiency.
1. An image decoding method performed by a decoding apparatus, the method comprising:
receiving image information through a bitstream;
generating luma prediction samples of a current block based on the image information; and
performing a reshaping process for the luma prediction samples,
wherein the performing the reshaping process includes:
deriving reshaper model indices for the reshaping; and
generating reshaped luma prediction samples based on the luma prediction samples and the reshaper model indices,
wherein the image information includes adaptation parameter set (APS),
wherein the APS includes reshaper model field,
wherein a reshaper model is derived based on the reshaper model field,
wherein the reshaper model indices are derived based on the reshaper model.
2. The method of claim 1 , wherein the image information includes header information,
wherein the header information includes a reshaper enabled flag and reshaper related APS ID information, and
wherein the reshaper related APS ID information represents an ID of the APS including the reshaper model field.
3. The method of claim 2 , wherein the header information includes a flag indicating whether to use the reshaper model field in the header information, and
wherein based on the flag indicating whether to use the reshaper model field in the header information, the reshaper model included in the APS indicated by the reshaper related APS ID information is used for the reshaping.
4. The method of claim 3 , wherein based on a determination that a value of the flag indicating whether to use the reshaper model field in the header information is 0, the reshaper model, which is derived based on the reshaper model field included in the APS indicated by the reshaper related APS ID information, is used for the reshaping, and
wherein based on a determination that the value of a flag indicating whether to use the reshaper model field in the header information is 1, an additional reshaper model field is included in the header information.
5. The method of claim 2 , wherein the method further comprises performing chroma residual scaling based on the reshaped prediction samples and a chroma residual scaling flag,
wherein the chroma residual scaling flag is included in the header information, and
When the chroma residual scaling flag is signaled based on a determination that a value of the reshaper enabled flag is 1.
6. The method of claim 2 , wherein the method further comprises:
generating reconstructed samples based on the reshaped luma prediction samples; and
performing an adaptive loop filter (ALF) process on the reconstructed samples,
wherein performing the ALF process comprises:
deriving filter coefficients for the ALF process; and
generating modified reconstructed samples based on the reconstructed samples and the filter coefficients,
wherein the APS includes ALF data, and
wherein the filter coefficients are derived by using the ALF data.
7. The method of claim 6 , wherein the header information includes an ALF enabled flag and ALF related APS ID information, and
wherein the ALF related APS ID information represents the ID of the APS including the ALF data.
8. The method of claim 7 , wherein the header information includes a flag indicating whether to use the ALF data in the header information,
wherein based on the flag indicating whether to use the ALF data in the header information, the ALF data included in the APS indicated by the ALF related APS ID information is used for deriving the filter coefficients.
9. The method of claim 8 , wherein based on a determination that a value of the flag indicating whether to use the ALF data in the header information is 0, the ALF data included in the APS indicated by the ALF related APS ID information is used for deriving the filter coefficients, and
an additional ALF data is included in the header information based on a determination that a value of a flag indicating whether to use the ALF data in the header information is 1.
10. An image encoding method performed by an encoding apparatus, the method comprising:
generating luma prediction samples of a current block in a current picture;
deriving reshaper model indices for reshaping on the luma prediction samples;
generating reshaped luma prediction samples based on the luma prediction samples and the reshaper model indices;
generating reshaping related information; and
encoding image information including prediction mode information for generating the luma prediction samples and the reshaping related information,
wherein the image information includes an adaptation parameter set (APS),
wherein the APS includes reshaper model field,
wherein a reshaper model is derived based on the reshaper model field,
wherein the reshaper model indices are derived based on the reshaper model.
11. The method of claim 10 , wherein the image information includes header information,
wherein the header information includes a reshaper enabled flag and reshaper related APS ID information, and
wherein the reshaper related APS ID information represents an ID of the APS including the reshaper model field.
12. The method of claim 11 , wherein the header information includes a flag indicating whether to use the reshaper model field in the header information, and
wherein based on the flag indicating whether to use the reshaper model field in the header information, the reshaper model included in the APS indicated by the reshaper related APS ID information is used for the reshaping.
13. The method of claim 12 , wherein based on a determination that a value of the flag indicating whether to use the reshaper model field in the header information is 0, the reshaper model, which is derived based on the reshaper model field included in the APS indicated by the reshaper related APS ID information, is used for the reshaping, and
wherein based on a determination that the value of a flag indicating whether to use the reshaper model field in the header information is 1, an additional reshaper model field is included in the header information.
14. The method of claim 11 , wherein the method further comprises:
generating reconstructed samples based on the reshaped luma prediction samples; and
deriving filter coefficients for an adaptive loop filter (ALF) process on the reconstructed samples;
generating modified reconstructed samples based on the reconstructed samples and the filter coefficients;
generating ALF data,
wherein the APS includes the ALF data, and
wherein the ALF data includes information for deriving the filter coefficients.
15. A non-transitory computer-readable digital storage medium, storing encoded information causing a decoding apparatus to perform an image decoding method, the method comprising: receiving image information through a bitstream; generating luma prediction samples of a current block based on the image information; and performing a reshaping process for the luma prediction samples, wherein the performing the reshaping process includes: deriving reshaper model indices for the reshaping; and generating reshaped luma prediction samples based on the luma prediction samples and the reshaper model indices, wherein the image information includes adaptation parameter set (APS), wherein the APS includes reshaper model field, wherein a reshaper model is derived based on the reshaper model field, wherein the reshaper model indices are derived based on the reshaper model.