Image decoding device, image encoding device, image processing system, and program

An image encoding device comprising a circuit, wherein

• the circuit:
• generates a prediction signal included in a prediction block based on a motion vector;
• and
• performs refinement processing of setting a search range based on a reference position specified by the motion vector, specifying a corrected reference position having the smallest predetermined cost from the search range, and correcting the motion vector based on the corrected reference position, when a predetermined condition is satisfied,
• the predetermined cost is a sum of absolute differences between a pixel in a reference block in the one reference frame and a pixel in a reference block in the other reference frame;
• in the refinement processing, the circuit:
• selects first candidate positions from the search range;
• selects at least one second candidate position from the first candidate positions and the reference position;
• selects only the reference position as the second candidate position when the predetermined cost of the reference position is smaller than a certain percentage of a minimum value of the predetermined cost of each of the first candidate positions, the certain percentage being less than 100% of the minimum value;
• specifies pixels adjacent to the second candidate position as third candidate positions, the third candidate position including half pixels interpolated by integer pixels;
• and
• specifies a position having the smallest predetermined cost from the first candidate positions, the second candidate position, the third candidate positions as the corrected reference position;
• the predetermined condition includes a condition that the prediction block is a block that performs bidirectional prediction, one reference frame is a frame that is temporally earlier than a target frame, and the other reference frame is a frame that is temporally future than the target frame;
• a displacement between the first candidate position and the reference position in the one reference frame is opposite to a sign of a displacement between the first candidate position and the reference position in the other reference frame;
• a displacement between the second candidate position and the reference position in the one reference frame opposite to a sign of a displacement between the second candidate position and the reference position in the other reference frame;
• a displacement between the third candidate position and the reference position in the one reference frame is opposite to a sign of a displacement between the third candidate position and the reference position in the other reference frame;
• and
• when a block size of the prediction block is larger than a predetermined block size, the circuit divides the prediction block into sub-block groups and performs the refinement processing for each sub-block.

An image processing system including an image encoding device and an image decoding device, wherein

• each of the image encoding device and the image decoding device include a circuit,
• the circuit:
• generates a prediction signal included in a prediction block based on a motion vector;
• performs refinement processing of setting a search range based on a reference position specified by the motion vector, specifying a corrected reference position having the smallest predetermined cost from the search range, and correcting the motion vector based on the corrected reference position, when a predetermined condition is satisfied;
• the predetermined cost is a sum of absolute differences between a pixel in a reference block in the one reference frame and a pixel in a reference block in the other reference frame;
• in the refinement processing, the circuit:
• selects first candidate positions from the search range;
• selects at least one second candidate position from the first candidate positions and the reference position;
• selects only the reference position as the second candidate position when the predetermined cost of the reference position is smaller than a certain percentage of a minimum value of the predetermined cost of each of the first candidate positions, the certain percentage being less than 100% of the minimum value;
• specifies pixels adjacent to the second candidate position as third candidate positions, the third candidate position including half pixels interpolated by integer pixels;
• and
• specifies a position having the smallest predetermined cost from the first candidate positions, the second candidate position, the third candidate positions as the corrected reference position;
• the predetermined condition includes a condition that the prediction block is a block that performs bidirectional prediction, one reference frame is a frame that is temporally earlier than a target frame, and the other reference frame is a frame that is temporally future than the target frame;
• a displacement between the first candidate position and the reference position in the one reference frame opposite to a sign of a displacement between the first candidate position and the reference position in the other reference frame;
• a displacement between the second candidate position and the reference position in the one reference frame is opposite to a sign of a displacement between the second candidate position and the reference position in the other reference frame;
• a displacement between the third candidate position and the reference position in the one reference frame is opposite to a sign of a displacement between the third candidate position and the reference position in the other reference frame;
• and
• when a block size of the prediction block is larger than a predetermined block size, the circuit divides the prediction block into sub-block groups and performs the refinement processing for each sub-block.

A non-transitory computer-readable medium that causes a computer to perform:

• a predicting step of generating a prediction signal based on a motion vector, wherein
• the predicting step includes a step of performing refinement processing of setting a search range based on a reference position specified by the motion vector, specifying a corrected reference position having the smallest predetermined cost from the search range, and correcting the motion vector based on the corrected reference position, when a predetermined condition is satisfied;
• the predetermined cost is a sum of absolute differences between a pixel in a reference block in the one reference frame and a pixel in a reference block in the other reference frame;
• in the step of performing refinement processing, the non-transitory computer-readable medium causes the computer to perform:
• selecting first candidate positions from the search range;
• selecting at least one second candidate position from the first candidate positions and the reference position;
• selecting only the reference position as the second candidate position when the predetermined cost of the reference position is smaller than a certain percentage of a minimum value of the predetermined cost of each of the first candidate positions, the certain percentage being less than 100% of the minimum value;
• specifying pixels adjacent to the second candidate position as third candidate positions, the third candidate position including half pixels interpolated by integer pixels;
• and
• specifying a position having the smallest predetermined cost from the first candidate positions, the second candidate position, the third candidate positions as the corrected reference position;
• the predetermined condition includes a condition that the prediction block is a block that performs bidirectional prediction, one reference frame is a frame that is temporally earlier than a target frame, and the other reference frame is a frame that is temporally future than the target frame;
• a displacement between the first candidate position and the reference position in the one reference frame is opposite to a sign of a displacement between the first candidate position and the reference position in the other reference frame;
• a displacement between the second candidate position and the reference position in the one reference frame is opposite to a sign of a displacement between the second candidate position and the reference position in the other reference frame;
• a displacement between the third candidate position and the reference position in the one reference frame is opposite to a sign of a displacement between the third candidate position and the reference position in the other reference frame;
• and
• in the step of performing refinement processing, when a block size of the prediction block is larger than a predetermined block size, the prediction block is divided into sub-block groups and the refinement processing is performed for each sub-block.

An image decoding method comprising:

• a predicting step of generating a prediction signal based on a motion vector, wherein
• the predicting step includes a step of performing refinement processing of setting a search range based on a reference position specified by the motion vector, specifying a corrected reference position having the smallest predetermined cost from the search range, and correcting the motion vector based on the corrected reference position, when a predetermined condition is satisfied;
• the predetermined cost is a sum of absolute differences between a pixel in a reference block in the one reference frame and a pixel in a reference block in the other reference frame;
• in the step of performing refinement processing, a non-transitory computer-readable medium causes a computer to perform:
• selecting first candidate positions from the search range;
• selecting at least one second candidate position from the first candidate positions and the reference position;
• selecting only the reference position as the second candidate position when the predetermined cost of the reference position is smaller than a certain percentage of a minimum value of the predetermined cost of each of the first candidate positions, the certain percentage being less than 100% of the minimum value;
• specifying pixels adjacent to the second candidate position as third candidate positions, the third candidate position including half pixels interpolated by integer pixels;
• and
• specifying a position having the smallest predetermined cost from the first candidate positions, the second candidate position, the third candidate positions as the corrected reference position;
• the predetermined condition includes a condition that the prediction block is a block that performs bidirectional prediction, one reference frame is a frame that is temporally earlier than a target frame, and the other reference frame is a frame that is temporally future than the target frame;
• a displacement between the first candidate position and the reference position in the one reference frame is opposite to a sign of a displacement between the first candidate position and the reference position in the other reference frame;
• a displacement between the second candidate position and the reference position in the one reference frame is opposite to a sign of a displacement between the second candidate position and the reference position in the other reference frame;
• a displacement between the third candidate position and the reference position in the one reference frame is opposite to a sign of a displacement between the third candidate position and the reference position in the other reference frame;
• and
• in the step of performing refinement processing, when a block size of the prediction block is larger than a predetermined block size, the prediction block is divided into sub-block groups and the refinement processing is performed for each sub-block.