Method of and apparatus for video encoding and decoding based on motion estimation
AssigneeSamsung Electronics Co., Ltd.
InventorsWoo-jin Han, Kyo-hyuk Lee, Tammy Lee
Life of this patent
Estimating the term…
Abstract
Provided is a method of and apparatus for video encoding and decoding based on motion estimation. The method includes generating a motion vector by searching a reference picture using pixels of a current block, generating a prediction motion vector that is a prediction value of the motion vector by searching the reference picture using previously encoded pixels located adjacent to the current block, and encoding the current block based on the motion vector and the prediction motion vector. By accurately predicting the motion vector of the current block, the number of bits required for encoding the motion vector can be reduced, thereby improving the compression rate of video data.
Claims21
A method of video encoding, the method comprising:
- generating a first motion vector by searching a reference picture using pixels of a current block;
- generating a prediction motion vector that is a prediction value of the first motion vector by searching the reference picture using previously encoded pixels located adjacent to the current block;
- generating a differential value between the first motion vector and the prediction motion vector;
- and
- encoding the current block based on the first motion vector and the prediction motion vector,
- wherein the encoding of the current block comprises encoding the differential value between the first motion vector and the prediction motion vector.
An apparatus for video encoding, the apparatus comprising:
- a motion vector generation unit which generates a first motion vector by searching a reference picture using pixels of a current block;
- a prediction motion vector generation unit which generates a prediction motion vector that is a prediction value of the first motion vector by searching the reference picture using previously encoded pixels located adjacent to the current block;
- a differential value generation unit which generates a differential value between the first motion vector and the prediction motion vector;
- and
- an encoding unit which encodes the current block based on the first motion vector and the prediction motion vector,
- wherein the encoding of the current block comprises encoding the differential value between the first motion vector and the prediction motion vector.
A method of video decoding, the method comprising:
- receiving a bitstream including data regarding a current block that has been encoded based on a first motion vector generated by searching a reference picture using pixels of the current block and a prediction motion vector that is a prediction value of the first motion vector and is generated by searching the reference picture using previously encoded pixels located adjacent to the current block;
- extracting data regarding the current block and data regarding a differential value between the prediction motion vector and the first motion vector;
- and
- performing motion compensation on the extracted data regarding the current block using the extracted data regarding the differential value motion vector,
- wherein the data regarding the differential value motion vector is a differential value between the first motion vector and the prediction motion vector.
An apparatus for video decoding, the apparatus comprising:
- a decoding unit which receives a bitstream including data regarding a current block that has been encoded based on a first motion vector generated by searching a reference picture using pixels of the current block and a prediction motion vector that is a prediction value of the first motion vector and is generated by searching the reference picture using previously encoded pixels located adjacent to the current block, and which extracts the data regarding the current block and data regarding a differential value between the prediction motion vector and the first motion vector from the received bitstream;
- and
- a motion compensation unit which performs motion compensation on the extracted data regarding the current block using the extracted data regarding the differential value motion vector,
- wherein the extracted data regarding the differential value motion vector is a differential value between the first motion vector and the prediction motion vector.
Dependencies are inferred from the claim text.