Method and apparatus for encoding image, and method and apparatus for decoding image

A method of encoding an image, the method comprising:

• classifying pixels of an input image and of a reference image into a first pixel group in which the bit depth is maintained, and a second pixel group in which the bit depth is changed;
• changing the bit depth of the second group pixels;
• generating a changed input image and a changed reference image by using the pixels in the first pixel group and also the pixels in the second pixel group, of which the bit depth has been changed;
• and
• performing motion prediction and compensation on the changed input image by using the changed reference image,
• wherein the classifying comprises:
• dividing the input image and the reference image into blocks having a predetermined size;
• and
• classifying at least one pixel at a certain location in each block as a member of the first pixel group, and classifying the remaining pixels in each block as members of the second pixel group.

An apparatus for encoding an image, the apparatus comprising at least one hardware processor which implements:

• a bit depth changing unit for classifying pixels of an input image into a first pixel group in which the bit depth is maintained, and a second pixel group in which the bit depth is changed, and changing the bit depth of second group pixels;
• a storage unit for storing a reference image used for motion prediction;
• a motion prediction unit for generating a motion vector by performing motion prediction by using the reference image and a changed input image generated by combining the pixels in the first pixel group and the pixels in the second pixel group;
• and
• a motion compensation unit for extracting a corresponding region of the reference image based on the motion vector,
• wherein the bit depth changing unit divides the input image into blocks having a predetermined size, classifies at least one pixel at a certain location in each block as a member of the first pixel group, and classifies the remaining pixels in each block as members of the second pixel group.

An apparatus for encoding an image, the apparatus comprising at least one hardware processor which implements:

• a bit depth changing unit for classifying pixels of an input image and of a reference image into a first pixel group in which the bit depth is maintained, and a second pixel group in which the bit depth is changed, and changing the bit depth of the second group pixels;
• a motion prediction unit for generating a motion vector by performing motion prediction using a changed input image and a changed reference image obtained by combining the pixels in the first pixel group and the pixels in the second pixel group;
• a storage unit for storing the changed reference image and the reference image;
• and
• a motion compensation unit for extracting a corresponding region of the reference image based on the motion vector,
• wherein the bit depth changing unit divides the input image into blocks having a predetermined size, classifies at least one pixel at a certain location in each block as a member of the first pixel group, and classifies the remaining pixels in each block as members of the second pixel group.

An image encoding method, comprising:

• dividing an input image into blocks of pixels having a predetermined size;
• for each of the pixel blocks, selecting at least one pixel at a certain location in each block for output at an unmodified bit depth;
• in each block, reducing the bit depth of the non-selected pixels;
• outputting, as a bitstream, a combination of the identified pixels at the unmodified bit depth and the non-selected pixels at the reduced bit depth.