Method and apparatus for applying view synthesized prediction according to illumination compensation
An inter-layer video decoding method is described including reconstructing a color image and a depth image of a first layer based on encoding information about the color image and the depth image of the first layer obtained from a bitstream. When it is determined that a prediction mode of a current block of a second layer image to be decoded is a view synthesized prediction mode based on an image synthesized from the first layer image a determination is made whether to perform luminance compensation on the second layer current block to be reconstructed. Use of the view synthesized prediction mode to reconstruct the second layer current block depends on the luminance compensation determination.
1. An inter-layer video decoding method implemented by a processor, the method comprising:
reconstructing, by the processor, a color image and a depth image of a first layer based on encoding information about the color image and the depth image of the first layer obtained from a bitstream;
when it is determined that a prediction mode of a current block of a second layer image to be decoded is a view synthesized prediction mode wherein prediction is performed based on an image synthesized from the first layer image, determining, by the processor, whether to perform luminance compensation on the second layer current block to be reconstructed; and
reconstructing, by the processor, the current block by determining to apply the view synthesized prediction mode to the current block only when it is determined not to perform the luminance compensation.
2. The inter-layer video decoding method of claim 1 , wherein
the determining of the prediction mode of the current block as the view synthesized prediction mode comprises determining, by the processor, the prediction mode of the current block as the view synthesized prediction mode when a block selected for predicting the current block from among blocks temporally and spatially adjacent to the current block is decoded in the view synthesized prediction mode.
3. The inter-layer video decoding method of claim 1 , further comprising:
determining, by the processor, a depth correspondence block of the first layer by inferring an initial disparity vector for the current block, from the block selected for predicting the current block from among the blocks temporally and spatially adjacent to the current block;
determining, by the processor, a depth value of the depth correspondence block and determining another disparity vector indicating a color correspondence block of the first layer with respect to the current block; and
predicting, by the processor, the current block by referring to the color correspondence block of the first layer that the other disparity vector indicates.
4. The inter-layer video decoding method of claim 1 , wherein
when it is determined to perform the luminance compensation, the method further comprises determining, by the processor, not to apply the view synthesized prediction mode to the current block, and
the inter-layer video decoding method further comprises:
determining, by the processor, a first layer reference block corresponding to the current block by inferring a disparity vector from the block selected for predicting the current block among the blocks temporally and spatially adjacent to the current block; and
reconstructing, by the processor, the current block by predicting the current block by using the first layer reference block.
5. The inter-layer video decoding method of claim 4 , further comprising, when it is determined to perform the luminance compensation, compensating, by the processor, for a luminance difference between the first layer reference block and the second layer current block with respect to the current block.
6. An inter-layer video encoding method implemented by a processor, the method comprising:
generating, by the processor, a bitstream including encoding information generated by encoding a color image and a depth image of a first layer;
when it is determined that a prediction mode of a current block of a second layer image to be encoded is a view synthesized prediction mode wherein a current block is predicted based on an image synthesized from the first layer, determining, by the processor, whether to perform luminance compensation on the second layer current block to be reconstructed; and
determining, by the processor, to apply the view synthesized prediction mode to the current block only when it is determined not to perform the luminance compensation.
7. The inter-layer video encoding method of claim 6 , wherein
the determining of the prediction mode of the current block as the view synthesized prediction mode comprises determining, by the processor, the prediction mode of the current block as the view synthesized prediction mode when a block selected for predicting the current block from among blocks temporally and spatially adjacent to the current block is decoded in the view synthesized prediction mode.
8. The inter-layer video encoding method of claim 6 , further comprising:
determining, by the processor, a depth correspondence block of the first layer by inferring an initial disparity vector for the current block, from the block selected for predicting the current block from among the blocks temporally and spatially adjacent to the current block;
determining, by the processor, a depth value of the depth correspondence block and determining another disparity vector indicating a color correspondence block of the first layer with respect to the current block; and
predicting, by the processor, the current block by referring to the color correspondence block of the first layer that the other disparity vector indicates.
9. The inter-layer video encoding method of claim 6 , wherein when it is determined to perform the luminance compensation, the method further comprises determining, by the processor, not to apply the view synthesized prediction mode to the current block, and
the inter-layer video encoding method further comprises:
determining, by the processor, a first layer reference block corresponding to the current block by inferring a disparity vector from the block selected for predicting the current block among the blocks temporally and spatially adjacent to the current block; and
reconstructing, by the processor, the current block by predicting the current block by using the first layer reference block.
10. The inter-layer video encoding method of claim 9 , further comprising, when it is determined to perform the luminance compensation, compensating, by the processor, for a luminance difference between the first layer reference block and the second layer current block with respect to the current block.
11. An inter-layer video decoding apparatus comprising:
a processor configured to reconstruct a color image and a depth image of a first layer based on encoding information about the color image and the depth image of the first layer obtained from a bitstream;
when it is determined that a prediction mode of a current block of a second layer image to be decoded is a view synthesized prediction mode wherein prediction is performed based on an image synthesized from a first layer image, determine whether to perform luminance compensation on the second layer current block to be reconstructed; and
reconstruct the current block by determining to apply the view synthesized prediction mode to the current block only when it is determined not to perform the luminance compensation.