Metadata-based image processing method and apparatus

1. A metadata generation method comprising:

acquiring at least one of a tone mapping function and a saturation correction function between a first image and a second image based on pixels included in the first image and pixels included in the second image;

correcting the first image by performing the at least one of the tone mapping function and the saturation correction function on the first image;

generating global metadata including the at least one of the tone mapping function and the saturation correction function;

splitting the corrected first image into a plurality of regions based on spatial proximity and color similarity of pixels included in the corrected first image;

acquiring a color mapping function of a first region among the plurality of regions by comparing color information of pixels in the first region with color information of pixels in the second image, which correspond to the pixels in the first region;

acquiring a color mapping function of a second region among the plurality of regions by comparing color information of pixels in the second region with color information of pixels in the second image, which correspond to the pixels in the second region;

setting the first region and the second region as a similar color mapping region by comparing the color mapping function of the first region with the color mapping function of the second region;

acquiring a color mapping function of the similar color mapping region based on the color mapping function of the first region and the color mapping function of the second region; and

generating local metadata including position information of the similar color mapping region in the corrected first image and the color mapping function of the similar color mapping region.

2. The metadata generation method of claim 1 , wherein the corrected first image is split into a foreground region and a background region based on a background model learned in a time axis, and

the plurality of regions are split from the foreground region and the background region.

3. The metadata generation method of claim 1 , wherein the first region and the second region are spatially adjacent.

4. The metadata generation method of claim 3 , further comprising:

acquiring a color mapping function of a third region, which is adjacent to the second region, among the plurality of regions by comparing color information of pixels in the third region with color information of pixels in the second image, which correspond to the pixels in the third region; and

setting the first region through the third region as the similar color mapping region or setting the third region as a new similar color mapping region, by comparing the color mapping function of the similar color mapping region with the color mapping function of the third region.

5. The metadata generation method of claim 1 , wherein the position information of the similar color mapping region comprises an upper left corner coordinate (x, y) and a right lower corner coordinate (x, y) of at least one bounding box with respect to the similar color mapping region.

6. The metadata generation method of claim 1 , wherein the position information of the similar color mapping region comprises at least one coordinate (x, y) of a bounding box and size information of the bounding box with respect to the similar color mapping region.

7. The metadata generation method of claim 1 , further comprising:

generating static metadata including gamut information of the second image; and

correcting a color gamut of the first image based on the gamut information of the second image,

wherein the acquiring of the at least one of the tone mapping function and the saturation correction function comprises:

acquiring the at least one of the tone mapping function and the saturation correction function based on pixels included in the gamut-corrected first image and pixels included in the second image.

8. The metadata generation method of claim 1 , further comprising:

generating static metadata including gamut information of the second image; and

correcting a color gamut of the first image based on the gamut information of the second image,

wherein the acquiring of the at least one of the tone mapping function and the saturation correction function comprises:

acquiring a tone mapping function between the gamut-corrected first image and the second image based on a luminance value of pixels included in the gamut-corrected first image and a luminance value of pixels included in the second image,

wherein the correcting of the first image by performing the at least one of the tone mapping function and the saturation correction function on the first image comprises:

performing tone mapping on the gamut-corrected first image based on the tone mapping function,

wherein the global metadata includes the tone mapping function, and

the plurality of regions are split from the tone-mapped first image.

9. The metadata generation method of claim 8 , wherein the acquiring of the at least one of the tone mapping function and the saturation correction function further comprises:

acquiring a saturation correction function between the tone-mapped first image and the second image based on a saturation value of pixels included in the tone-mapped first image and a saturation value of the pixels included in the second image; and

performing saturation correction on the tone-mapped first image based on the saturation correction function,

wherein the global metadata further comprises the saturation correction function, and

the plurality of regions are split from the saturation-corrected first image.

10. An image processing method comprising:

receiving global metadata including at least one of a tone mapping function and a saturation correction function between a first image and a second image;

correcting the first image by performing the at least one of the tone mapping function and the saturation correction function on the first image;

receiving local metadata including position information of a similar color mapping region in the corrected first image and a color mapping function of the similar color mapping region;

restoring the similar color mapping region in the corrected first image based on the position information of the similar color mapping region; and

correcting color components of pixels in the restored similar color mapping region based on the color mapping function of the similar color mapping region.

11. The image processing method of claim 10 , wherein the position information of the similar color mapping region comprises an upper left corner coordinate (x, y) and a right lower corner coordinate (x, y) of a bounding box with respect to the similar color mapping region.

12. The image processing method of claim 10 , wherein the position information of the similar color mapping region comprises at least one coordinate (x, y) of a bounding box and size information of the bounding box with respect to the similar color mapping region.

13. The image processing method of claim 10 , further comprising:

receiving static metadata including gamut information of the second image; and

correcting a color gamut of the first image based on the gamut information of the second image,

wherein the correcting of the first image comprises correcting the gamut-corrected first image by performing the at least one of the tone mapping function and the saturation correction function on the gamut-corrected first image.

14. The image processing method of claim 10 , further comprising:

receiving static metadata including gamut information of the second image; and

correcting a color gamut of the first image based on the gamut information of the second image,

wherein the receiving of the global metadata comprises receiving the global metadata including the tone mapping function between the gamut-corrected first image and the second image,

wherein the correcting of the first image by performing the at least one of the tone mapping function and the saturation correction function comprise:

performing tone mapping on the gamut-corrected first image based on the tone mapping function, and

the similar color mapping region is restored from the tone-mapped first image.

15. The image processing method of claim 14 , wherein the correcting of the first image by performing the at least one of the tone mapping function and the saturation correction function further comprises performing saturation correction on the tone-mapped first image based on a saturation correction function between the tone-mapped first image and the second image, which is included in the global metadata,

wherein the similar color mapping region is restored from the saturation-corrected first image.

16. An image decoding apparatus comprising:

a communicator configured to receive a first image, local metadata, and global metadata; and

a processor configured to:

receive global metadata including at least one of a tone mapping function and a saturation correction function between the first image and a second image,

correct the first image by performing the at least one of the tone mapping function and the saturation correction function on the first image,

receive the local metadata including position information of a similar color mapping region in the corrected first image and a color mapping function of the similar color mapping region,

restore the similar color mapping region in the corrected first image based on the position information of the similar color mapping region; and

correct color components of pixels in the restored similar color mapping region based on the color mapping function of the similar color mapping region.

17. The image decoding apparatus of claim 16 , wherein the position information of the similar color mapping region comprises an upper left corner coordinate (x, y) and a right lower corner coordinate (x, y) of a bounding box with respect to the similar color mapping region.

18. The image decoding apparatus of claim 16 , wherein the position information of the similar color mapping region comprises at least one coordinate (x, y) of a bounding box and size information of the bounding box with respect to the similar color mapping region.

19. The image decoding apparatus of claim 16 , wherein the processor configured to correct a color gamut of the first image based on static metadata including gamut information of the second image, which has been received by the communicator, and correct the gamut-corrected first image by performing the at least one of the tone mapping function and the saturation correction function on the gamut-corrected first image.

20. The image decoding apparatus of claim 19 , wherein the processor configured to perform tone mapping on the gamut-corrected first image based on global metadata including the tone mapping function between the gamut-corrected first image and the second image, which has been received by the communicator,

wherein the similar color mapping region is restored from the tone-mapped first image.