Image processing device and method
Inverse quantization is performed on a quantization coefficient using a quantization parameter depending on whether a transform skip is to be applied, inverse coefficient transformation is performed on a transform coefficient generated by inverse quantization to generate a predicted residual that is a residual between an image and a predicted image of the image when the transform skip is not to be applied, and the inverse coefficient transformation is skipped when the transform skip is to be applied. This can be applied to, for example, an image processing device, an image coding device, an image decoding device, a transmission device, a reception device, a transmission/reception device, an information processing device, an imaging device, a reproduction device, an electronic apparatus, an image processing method, an information processing method, or the like.
1. An image processing device comprising:
at least one processor; and
at least one memory including computer program, wherein the at least one memory and the computer program are configured, with the at least one processor, to cause the image processing device to at least
perform inverse quantization on a quantization coefficient;
normalize a transform coefficient generated by the inverse quantization using a scaling parameter depending on whether a transform skip is to be applied, wherein the transform coefficient is normalized using the scaling parameter that reflects the decompression processing when the transform skip is to be applied, and
perform inverse coefficient transformation on the transform coefficient normalized to generate a predicted residual that is a residual between an image and a predicted image of the image when the transform skip is not to be applied, and skip the inverse coefficient transformation without performing the decompression processing when the transform skip is applied.
2. The image processing device according to claim 1 , wherein
the transform coefficient is normalized using the scaling parameter that does not reflect decompression processing in the transform skip when the transform skip is not to be applied, and
the inverse coefficient transformation is performed on the transform coefficient normalized using the scaling parameter that does not reflect the decompression processing when the transform skip is not to be applied.
3. The image processing device according to claim 2 , wherein
a component determined on the basis of a transform block size of the scaling parameter is set to “0” when the transform skip is to be applied.
4. The image processing device according to claim 2 , wherein
a component determined on the basis of the transform block size of the scaling parameter is set to a value according to a quantization matrix when the transform skip is to be applied.
5. The image processing device according to claim 4 , wherein
the value according to the quantization matrix is “−4”.
6. An image processing method comprising:
performing inverse quantization on a quantization coefficient;
normalizing a transform coefficient generated by the inverse quantization using a scaling parameter depending on whether transform skip is to be applied, wherein the transform coefficient is normalized using the scaling parameter that reflects decompression processing in the transform skip when the transform skip is to be applied; and
performing inverse coefficient transformation on the normalized transform coefficient to generate a predicted residual that is a residual between an image and a predicted image of the image when the transform skip is not to be applied and skipping the inverse coefficient transformation without performing the decompression processing when the transform skip is to be applied.
7. An image processing device comprising:
at least one processor; and
at least one memory including computer program, wherein the at least one memory and the computer program are configured, with the at least one processor, to cause the image processing device to at least
perform coefficient transformation on a predicted residual that is a residual between an image and a predicted image of the image to generate a transform coefficient when a transform skip is not to be applied, and skip the coefficient transformation without performing decompression processing when the transform skip is to be applied;
perform quantization on the transform coefficient when the transform skip is not to be applied and perform the quantization on the predicted residual when the transform skip is to be applied; and
normalize a quantization coefficient generated using a scaling parameter depending on whether the transform skip is to be applied, wherein the quantization coefficient is normalized using the scaling parameter reflecting the decompression processing when the transform skip is to be applied.
8. The image processing device according to claim 7 , wherein
the coefficient transformation is performed on the predicted residual when the transform skip is not to be applied, and
wherein the quantization coefficient is normalized using the scaling parameter that does not reflect the decompression processing in the transform skip when the transform skip is not to be applied.
9. The image processing device according to claim 8 , wherein
a component determined on the basis of the transform block size of the scaling parameter is set to a value according to a quantization matrix when the transform skip is to be applied.
10. The image processing device according to claim 9 , wherein
the value according to the quantization matrix is “−4”.
11. An image processing method comprising:
performing coefficient transformation on a predicted residual that is a residual between an image and a predicted image of the image to generate a transform coefficient when a transform skip is not to be applied and skipping the coefficient transformation without performing decompression processing when the transform skip is to be applied;
performing quantization on the transform coefficient when the transform skip is not to be applied and performing the quantization on the predicted residual when the transform skip is to be applied; and
normalizing a quantization coefficient generated by the quantization using a scaling parameter depending on whether the transform skip is to be applied, wherein the quantization coefficient is normalized using the scaling parameter reflecting the decompression processing when the transform skip is to be applied.
12. An image processing device comprising:
at least one processor; and
at least one memory including computer program, wherein the at least one memory and the computer program are configured, with the at least one processor, to cause the image processing device to at least
perform inverse quantization on a quantization coefficient;
normalize a transform coefficient generated by the inverse quantization using a scaling parameter depending on whether a transform skip is to be applied,
perform inverse coefficient transformation on the transform coefficient normalized to generate a predicted residual that is a residual between an image and a predicted image of the image when the transform skip is not to be applied, and skip the inverse coefficient transformation when the transform skip is applied, and
wherein the transform coefficient is normalized using the scaling parameter reflecting normalization processing for the predicted residual when the transform skip is to be applied.
13. The image processing device according to claim 12 , wherein the at least one memory and the computer program are configured, with the at least one processor, to cause the image processing device to at least
normalize the predicted residual when the transform skip is not to be applied.
14. The image processing device according to claim 13 , wherein
the transform coefficient is normalized using the scaling parameter including a value according to inverse quantization and a scaling list and a value according to application of inverse-dependence quantization when the transform skip is to be applied.
15. The image processing device according to claim 14 , wherein
the value according to inverse quantization and the scaling list is “10”.
16. An image processing method comprising:
performing inverse quantization on a quantization coefficient;
normalizing a transform coefficient generated by the inverse quantization using a scaling parameter depending on whether transform skip is to be applied;
performing inverse coefficient transformation on the normalized transform coefficient to generate a predicted residual that is a residual between an image and a predicted image of the image when the transform skip is not to be applied and skipping the inverse coefficient transformation when the transform skip is to be applied,
wherein the quantization coefficient is normalized using the scaling parameter reflecting normalization processing for the predicted residual when the transform skip is to be applied.
17. An image processing device comprising:
at least one processor; and
at least one memory including computer program, wherein the at least one memory and the computer program are configured, with the at least one processor, to cause the image processing device to at least
perform coefficient transformation on a predicted residual that is a residual between an image and a predicted image of the image to generate a transform coefficient when a transform skip is not to be applied, and skip the coefficient transformation when the transform skip is to be applied;
perform quantization on the transform coefficient when the transform skip is not to be applied and perform the quantization on the predicted residual when the transform skip is to be applied; and
normalize a quantization coefficient generated by the quantization using a scaling parameter depending on whether the transform skip is to be applied, wherein the quantization coefficient is normalized using the scaling parameter reflecting normalization processing for the transform coefficient when the transform skip is to be applied.
18. The image processing device according to claim 17 , wherein the at least one memory and the computer program are configured, with the at least one processor, to cause the image processing device to at least
normalize the transform coefficient when the transform skip is not to be applied.
19. The image processing device according to claim 18 , wherein
the transform coefficient is normalized using the scaling parameter including a value according to quantization and a scaling list and a value according to application of dependent quantization when the transform skip is to be applied.
20. The image processing device according to claim 19 , wherein
the value according to quantization and the scaling list is “10”.
21. An image processing method comprising:
performing coefficient transformation on a predicted residual that is a residual between an image and a predicted image of the image to generate a transform coefficient when a transform skip is not to be applied and skipping the coefficient transformation when the transform skip is to be applied;
performing quantization on the transform coefficient when the transform skip is not to be applied and performing the quantization on the predicted residual when the transform skip is to be applied; and
normalizing a quantization coefficient generated by the quantization using a scaling parameter depending on whether the transform skip is to be applied, wherein the quantization coefficient is normalized using the scaling parameter reflecting a normalization processing for the transform coefficient when the transform skip is to be applied.