Prediction image generation device, video decoding device, and video coding device
At least one of a vector candidate derivation unit and a merge candidate derivation unit derives a motion vector of each of multiple sub-blocks contained in a decoding target block by referring to motion vectors at multiple control points including two points at an interval which is longer than one side of the target block, and a prediction image generation unit refers to the motion vector of each sub-block to generate a prediction image.
1. A prediction image generation device for generating a prediction image used to code or decode a video, the prediction image generation device comprising:
a prediction vector computing circuit; and
a prediction image generation circuit, wherein
the prediction vector computing circuit derives at least one control point motion vector by using a neighboring location and a motion vector in a neighboring block adjacent to a target block,
the prediction vector computing circuit derives a subblock motion vector by using a sum of a difference motion vector and a control point motion vector predictor, and
the prediction image generation circuit generates the prediction image by referring to the subblock motion vector,
wherein the neighboring location is derived by using one of (i) a first neighboring location (xPb−1, yPb−1), (ii) a second neighboring location (xPb+W−1, yPb−1), (iii) a third neighboring location (xPb+W, yPb−1), (iv) a fourth neighboring location (xPb−1, yPb+H−1) and (v) a fifth neighboring location (xPb−1, yPb+H), wherein xPb is an X coordinate of the target block, yPb is a Y coordinate of the target block, H is a height of the target block and W is a width of the target block.
2. A video decoding device comprising
the prediction image generation device according to claim 1 , wherein
the video decoding device reconstructs a coding target image by adding a residual image to the prediction image.
3. A video coding device comprising
the prediction image generation device according to claim 1 , wherein
the video coding device codes a residual error between the prediction image and a coding target image.