Motion estimation apparatus and method and image encoding apparatus and method employing the same
AssigneeSamsung Electronics Co., Ltd.
InventorsSang-jo Lee, Ho-jin Lee, Do-hyung Kim
Life of this patent
Estimating the term…
Abstract
Provided are a motion estimation apparatus and method and an image encoding apparatus and method employing the same. The motion estimation apparatus includes an optimal motion estimation unit performing motion estimation in an initial block mode while skipping remaining block modes excluding the initial block mode from a plurality of block modes of the current block, or performing motion estimation in candidate block modes determined from the plurality of block modes.
Claims23
A motion estimation apparatus comprising:
- a first motion estimation unit, implemented by at least one processing device, to perform a first motion estimation on a current block using an initial motion vector in an initial block mode and to determine a block mode of the current block based on a result of the first motion estimation;
- and
- a second motion estimation unit to determine at least one candidate block mode including the initial block mode among a plurality of block modes according to motion characteristics of pixels included in the current block and to perform a second motion estimation on the current block in either the initial block mode or the at least one candidate block mode so as to determine a motion vector of the current block,
- wherein the second motion estimation unit comprises:
- a candidate block mode determination unit to determine the at least one candidate block mode of the current block according to a result of comparing a respective plurality of motion vectors of a plurality of sub-blocks forming the current block with each other;
- and
- a block mode skip determination unit to determine whether to skip remaining block modes excluding the initial block mode from the at least one candidate block mode of the current block after determining the at least one candidate block mode of the current block.
A motion estimation apparatus comprising:
- a first motion estimation unit, implemented by at least one processing device, to perform a first motion estimation on a current block using an initial motion vector in an initial block mode and to determine the initial block mode to be a block mode of the current block according to a result of the first motion estimation;
- a candidate block mode determination unit to determine at least one candidate block mode, including the initial block mode among a plurality of block modes, of the current block according to motion characteristics of pixels included in the current block by comparing a respective plurality of motion vectors of a plurality of sub-blocks forming the current block with each other;
- a block mode skip determination unit to determine whether to skip remaining block modes excluding the initial block mode from the at least one candidate block mode of the current block after determining the at least one candidate block mode of the current block;
- and
- a motion vector determination unit to determine a motion vector of the current block using either the initial block mode or the at least one candidate block mode based on a result of determination in the block mode skip determination unit.
A motion estimation method comprising:
- performing, by using at least one processing device, a first motion estimation on a current block using an initial motion vector in an initial block mode to determine a block mode of the current block based on a result of the first motion estimation;
- and
- performing a second motion estimation on the current block in either the initial block mode or at least one candidate block mode including the initial block mode among a plurality of block modes by determining the at least one candidate block mode according to motion characteristics of pixels included in the current block,
- wherein the performing of the second motion estimation comprises:
- determining the at least one candidate block mode of the current block according to a result of comparing a respective plurality of motion vectors of a plurality of sub-blocks forming the current block with each other;
- and
- determining whether to skip remaining block modes excluding the initial block mode from the at least one candidate block mode of the current block after determining the at least one candidate block mode of the current block.
A motion estimation method comprising:
- performing, by using at least one processing device, a first motion estimation on a current block using an initial motion vector in an initial block mode and determining the initial block mode to be a block mode of the current block according to a result of the first motion estimation;
- determining at least one candidate block mode, including the initial block mode among a plurality of block modes, of the current block according to motion characteristics of pixels included in the current block by comparing a respective plurality of motion vectors of a plurality of sub-blocks forming the current block with each other;
- determining whether to skip remaining block modes excluding the initial block mode from the at least one candidate block mode of the current block after determining the at least one candidate block mode of the current block;
- and
- determining a motion vector of the current block using either the initial block mode or the at least one candidate block mode, based on a result of determining whether to skip remaining block modes.
An image encoding apparatus comprising:
- a motion estimation unit, implemented by at least one processing device, to perform a first motion estimation on a current block using an initial motion vector in an initial block mode, to determine a block mode of the current block based on a result of the first motion estimation, to determine at least one candidate block mode including the initial block mode among a plurality of block modes according to motion characteristics of pixels included in the current block by comparing a respective plurality of motion vectors of a plurality of sub-blocks forming the current block with each other, to determine whether to skip remaining block modes excluding the initial block mode from the at least one candidate block mode of the current block after determining the at least one candidate block mode of the current block and to perform a second motion estimation on the current block in either the initial block mode or the at least one candidate block mode;
- and
- an encoding unit to encode a residual signal generated by using an output of the motion estimation unit.
An image encoding method comprising:
- performing, by using at least one processing device, a first motion estimation on a current block using an initial motion vector in an initial block mode, determining a block mode of the current block based on a result of the first motion estimation, determining at least one candidate block mode including the initial block mode among a plurality of block modes according to motion characteristics of pixels included in the current block by comparing a respective plurality of motion vectors of a plurality of sub-blocks forming the current block with each other, determining whether to skip remaining block modes excluding the initial block mode from the at least one candidate block mode of the current block after determining the at least one candidate block mode of the current block and performing a second motion estimation on the current block in either the initial block mode or the at least one candidate block mode;
- and
- encoding a residual signal provided by the performing of the motion estimation.
Dependencies are inferred from the claim text.