Image processing apparatus and method
The present disclosure relates to image processing apparatus and method that can suppress an increase in a load of encoding and decoding. One or both a predetermined upper limit and a predetermined lower limit are used to clip transformation coefficients obtained by applying a transformation process to a predicted residual that is a difference between an image and a predicted image of the image. In addition, one or both a predetermined upper limit and a predetermined lower limit are used to clip transformation coefficients that are subjected to an inverse transformation process to obtain a predicted residual that is a difference between an image and a predicted image of the image. The present disclosure can be applied to, for example, an image processing apparatus, an image encoding apparatus, an image decoding apparatus, and the like.
1. An image processing apparatus comprising:
a clip processing section configured to use one or both a predetermined upper limit and a predetermined lower limit to clip transformation coefficients obtained by applying a transformation process to a predicted residual that is a difference between an image and a predicted image of the image,
wherein the predetermined upper limit and the predetermined lower limit are each decided based on a bit depth of an input signal including the transformation coefficients,
wherein the predetermined upper limit and the predetermined lower limit are each further decided based on an extended computation precision flag of the input signal, and
wherein the clip processing section is implemented via at least one processor.
2. The image processing apparatus according to claim 1 , further comprising:
a primary transformation section configured to perform a primary transformation that is a transformation process for the predicted residual; and
a secondary transformation section configured to perform a secondary transformation that is a transformation process for primary transformation coefficients obtained by the primary transformation of the predicted residual performed by the primary transformation section,
wherein the clip processing section is further configured to clip secondary transformation coefficients obtained by the secondary transformation of the primary transformation coefficients performed by the secondary transformation section, and
wherein the primary transformation section and the secondary transformation section are each implemented via at least one processor.
3. The image processing apparatus according to claim 1 , further comprising:
a rasterization section configured to transform the transformation coefficients into a one-dimensional vector;
a matrix computation section configured to perform matrix computation of the one-dimensional vector;
a scaling section configured to scale the one-dimensional vector subjected to the matrix computation; and
a matrix formation section configured to form a matrix of the scaled one-dimensional vector,
wherein the clip processing section is further configured to clip the one-dimensional vector scaled by the scaling section,
wherein the matrix formation section is further configured to form a matrix of the one-dimensional vector clipped by the clip processing section, and
wherein the rasterization section, the matrix computation section, the scaling section, and the matrix formation section are each implemented via at least one processor.
4. An image processing method comprising:
using one or both a predetermined upper limit and a predetermined lower limit to clip transformation coefficients obtained by applying a transformation process to a predicted residual that is a difference between an image and a predicted image of the image,
wherein the predetermined upper limit and the predetermined lower limit are each decided based on a bit depth of an input signal including the transformation coefficients, and
wherein the predetermined upper limit and the predetermined lower limit are each further decided based on an extended computation precision flag of the input signal.