Image encoding/decoding method and apparatus, and recording medium storing bitstream
An image encoding/decoding method is provided. An image decoding method of the present invention may comprise restoring an intra-prediction mode of a current chroma block, deriving at least one representative value based on a neighboring sample of the current chroma block and a neighboring sample of a corresponding luma block corresponding to the current chroma block in case the intra-prediction mode is a CCLM (Cross-Component Linear Model) mode, deriving a parameter of CCLM by using the at least one representative value, and generating a prediction block of the current chroma block by using the parameter of CCLM.
1. A method of decoding an image, the method comprising:
restoring an intra-prediction mode of a current chroma block;
deriving at least one representative value based on a neighboring sample of the current chroma block and a neighboring sample of a corresponding luma block corresponding to the current chroma block in case the intra-prediction mode is a CCLM (Cross-Component Linear Model) mode;
deriving a parameter of CCLM by using the at least one representative value; and
generating a prediction block of the current chroma block by using the parameter of CCLM,
wherein the deriving of at least one representative value comprises
deriving positions of N neighboring samples;
deriving N candidate values of luma components from neighboring samples of the luma block based on the derived positions;
deriving N candidate values of chroma components from neighboring samples of the current chroma block based on the derived positions;
deriving a representative value of a luma component based on the N candidate values of the luma components; and
deriving a representative value of a chroma component based on the N candidate values of the chroma components.
2. The method of claim 1 ,
wherein the N is determined based on the number of available neighboring samples and the intra-prediction mode.
3. The method of claim 1 ,
wherein positions of the N neighboring samples are derived at an equal spacing.
4. The method of claim 3 ,
wherein the equal spacing is determined based on the number of available neighboring samples and the intra-prediction mode.
5. The method of claim 1 ,
wherein the position of a first neighboring sample among the positions of the N neighboring samples is determined based on the number of available neighboring samples and the intra-prediction mode.
6. The method of claim 1 ,
wherein the deriving of a candidate value of a luma component comprises
selecting M neighboring samples from neighboring samples of the luma block based on the derived position; and
deriving a statistical value of the M neighboring samples as the candidate value.
7. The method of claim 6 ,
wherein the statistical value is a weighted average value of the M neighboring samples.
8. The method of claim 6 , wherein the M is 5 or 6.
9. The method of claim 1 ,
wherein the N is 4,
the deriving of a representative value of a luma component
derives an average of a first maximum value and a second maximum value among the four candidate values of luma components as a first representative value, and
derives an average of a first minimum value and a second minimum value among the four candidate values of luma components as a third representative value, and
the deriving of a representative value of a chroma component
derives an average of a first maximum value and a second maximum value among the four candidate values of chroma components as a second representative value, and
derives an average of a first minimum value and a second minimum value among the four candidate values of chroma components as a fourth representative value.
10. A method of encoding an image, the method comprising:
determining an intra-prediction mode of a current chroma block;
deriving at least one representative value based on a neighboring sample of the current chroma block and a neighboring sample of a corresponding luma block corresponding to the current chroma block in case the intra-prediction mode is a CCLM (Cross-Component Linear Model) mode;
deriving a parameter of CCLM by using the at least one representative value; and
generating a prediction block of the current chroma block by using the parameter of CCLM,
wherein the deriving of at least one representative value comprises
deriving positions of N neighboring samples;
deriving N candidate values of luma components from neighboring samples of the luma block based on the derived positions;
deriving N candidate values of chroma components from neighboring samples of the current chroma block based on the derived positions;
deriving a representative value of a luma component based on the N candidate values of the luma components; and
deriving a representative value of a chroma component based on the N candidate values of the chroma components.
11. The method of claim 10 ,
wherein the N is determined based on the number of available neighboring samples and the intra-prediction mode.
12. The method of claim 10 ,
wherein positions of the N neighboring samples are derived at an equal spacing.
13. The method of claim 12 ,
wherein the equal spacing is determined based on the number of available neighboring samples and the intra-prediction mode.
14. The method of claim 10 ,
wherein the position of a first neighboring sample among the positions of the N neighboring samples is determined based on the number of available neighboring samples and the intra-prediction mode.
15. The method of claim 10 ,
wherein the deriving of a candidate value of a luma component comprises
selecting M neighboring samples from neighboring samples of the luma block based on the derived position; and
deriving a statistical value of the M neighboring samples as the candidate value.
16. The method of claim 15 ,
wherein the statistical value is a weighted average value of the M neighboring samples.
17. The method of claim 10 ,
wherein the N is 4,
the deriving of a representative value of a luma component
derives an average of a first maximum value and a second maximum value among the four candidate values of luma components as a first representative value, and
derives an average of a first minimum value and a second minimum value among the four candidate values of luma components as a third representative value, and
the deriving of a representative value of a chroma component
derives an average of a first maximum value and a second maximum value among the four candidate values of chroma components as a second representative value, and
derives an average of a first minimum value and a second minimum value among the four candidate values of chroma components as a fourth representative value.
18. A non-transitory computer-readable recording medium storing a bitstream which is received, decoded and used by an image decoding apparatus to reconstruct an image, wherein the bitstream comprises information on an intra-prediction mode of a current chroma block,
information on an intra-prediction mode is used to restore an intra-prediction mode of the current chroma block;
at least one representative value is derived based on a neighboring sample of the current chroma block and a neighboring sample of a corresponding luma block corresponding to the current chroma block in case the intra-prediction mode is a CCLM (Cross-Component Linear Model) mode,
the at least one representative value is used to derive a parameter of CCLM, and
the parameter of CCLM is used to generate a prediction block of the current chroma block, and
wherein when at least one representative value is derived, positions of N neighboring samples are derived, N candidate values of luma components from neighboring samples of the luma block are derived based on the derived positions, N candidate values of chroma components from neighboring samples of the current chroma block are derived based on the derived positions, a representative value of a luma component is derived based on the N candidate values of the luma components, and a representative value of a chroma component is derived based on the N candidate values of the chroma components.