Method and device for encoding/decoding image by using geometrically changed image
A method and apparatus use a geometric modified image for video encoding/decoding. The encoding method may include: generating a geometric modified reference picture by geometrically modifying a reference picture; generating a prediction block of a current block within an encoding target picture by performing inter prediction by referencing the reference picture or the geometrically modified reference picture; and encoding inter-prediction information of the current block.
1. A method for encoding an image, the method comprising:
determining a prediction block of a current block within an encoding target picture by performing inter prediction based on a reference picture and information about geometric modification; and
encoding geometric modification usage information of the current block indicating whether geometric modification information is used for inter-prediction for a current block,
wherein geometric modification for a part of the reference picture according to the geometric modification information includes at least one of a size modification, and a rotation modification, and
the prediction block is determined based on a geometrically modified part of the reference picture according to the geometric modification information.
2. The method of claim 1 , further comprising:
generating geometric modification information based on a relation between the encoding target picture and the reference picture, wherein the geometric modified reference picture is generated based on the geometric modified information.
3. The method of claim 1 , wherein the prediction block is selected from either a first prediction block generated by inter prediction referencing the reference picture or a second prediction block generated by inter prediction based on the reference picture and the geometric modification information.
4. The method of claim 3 , wherein the selected prediction block is selected based on an encoding efficiency of the current block for each of the first prediction block and the second prediction block.
5. The method of claim 4 , wherein the encoding efficiency of the current block is determined based on a rate-distortion cost.
6. The method of claim 1 , wherein the geometric modification usage information is encoded in a symbol value corresponding to a combination of information on each prediction direction and geometric modification usage information on each prediction direction when a bi or more-directional prediction is used for generating the prediction block of the current block.
7. The method of claim 6 , wherein encoding the symbol value is performed by encoding a difference value between the symbol value and a symbol value corresponding to geometric modification usage information that is previously used.
8. A method for decoding an image, the method comprising:
decoding geometric modification usage information of a current block indicating whether information about geometric modification is used for inter-prediction for a current block; and
when the geometric modification usage information indicates that the information about geometric modification is used for inter prediction for the current block, determining a prediction block of the current block within a decoding target picture by performing inter prediction based on a reference picture and the geometric modification information,
wherein geometric modification for a part of the reference picture according to the geometric modification information includes at least one of a size modification, and a rotation modification, and
the prediction block is determined based on a geometrically modified part of the reference picture according to the geometric modification.
9. The method of claim 8 , wherein when the geometric modification usage information indicates that the geometric modified reference picture is used, the decoding method further comprises
decoding geometric modification information; and
generating the geometric modified reference picture by geometrically modifying the reference picture based on the geometric modification usage information, and
wherein the prediction block of the current block is generated by inter prediction referencing the geometric modified reference picture.
10. The method of claim 8 , wherein the geometric modification usage information is decoded in a symbol value corresponding to a combination of information of each prediction direction and geometric modification usage information on each prediction direction when a bi or more-directional prediction is used for generating the prediction block of the current block.
11. The method of claim 10 , wherein decoding the symbol value comprises
decoding a difference value between the symbol value and a symbol value corresponding to geometric modification usage information that is previously used; and
adding the symbol value corresponding to geometric modification usage information that is previously used to the decoded difference value.
12. The method of claim 9 , wherein the geometric modification information is generated based on a relation between the decoding target picture and the reference picture, and has various forms including at least one of a global motion vector, a geometric transfer modification matrix, a geometric size modification matrix, a geometric affine modification matrix, and a geometric projection modification matrix.
13. An apparatus for encoding an image, the apparatus comprising:
an inter-prediction unit configured for determining a prediction block of a current block within an encoding target picture by performing inter prediction based on a reference picture and information about geometric modification; and
an encoder configured for encoding geometric modification usage information of the current block indicating whether geometric modification information is used for inter-prediction for a current block,
wherein geometric modification for a part of the reference picture according to the geometric modification information includes at least one of a size modification, and a rotation modification, and
the prediction block is determined based on a geometrically modified part of the reference picture according to the geometric modification information.
14. An apparatus for decoding an image, the apparatus comprising:
a decoder configured for decoding geometric modification usage information of a current block indicating whether information about geometric modification is used for inter-prediction for a current block; and
an inter-prediction unit configured for, when the geometric modification usage information indicates that the information about geometric modification is used for inter prediction for the current block, determining a prediction block of the current block within a decoding target picture by performing inter prediction based on a reference picture and the geometric modification information,
wherein geometric modification for a part of the reference picture according to the geometric modification information includes at least one of a size modification, and a rotation modification, and
the prediction block is determined based on a geometrically modified part of the reference picture according to the geometric modification information.