Video encoding device and video decoding device
In a case of referring to information between pictures by means of inter prediction, the entire picture needs to be decoded even in a case of decoding a small region. In a case of referring to information within a picture by means of intra prediction, referring to information across a tile boundary is not possible. In a case that a pixel pointed by a sub-block level motion vector of a target block that is calculated by scaling based on an available motion vector acquired from a spatial neighboring block or a temporal neighboring block is not present within a tile sequence, a process of replacing the pixel value with a pixel value within the tile sequence is performed. In a case of referring to a pixel outside of a tile by means of intra prediction, a process of replacing the pixel value with that of a pixel within the tile is performed.
1. A video decoding device for generating a prediction image from two motion vectors, the video decoding device comprising:
a memory and
a processor, wherein the processor is configured to:
decode a flag indicating whether or not a current tile, obtained by splitting a current picture into rectangular regions and including a current block, is independently decoded;
derive two motion vectors of the current block;
in a case that a first one of the two motion vectors points to a first reference picture which is located prior to a current picture in a time direction and a second one of the two motion vectors points to a second reference picture which is located after the current picture in the time direction,
a) perform a motion vector search around each of the two motion vectors by using a range value;
b) derive reference positions in each of the first reference picture and the second reference picture based on upper left coordinates of the current block, each of the two motion vectors, and the range value;
c) derive clipped positions by clipping x coordinates of the reference positions in a range from xTs to xTs+wT−1 and clipping y coordinates of the reference positions in a range from yTs to yTs+hT−1 in a case that the flag indicates the current tile is independently decoded, where the xTs is an
x coordinate of upper left coordinates of the current tile, where the wT is a width of the current tile, where the yTs is a y coordinate of the upper left coordinates of the current tile, and where the hT is a height of the current tile;
d) generate compensation images based on pixel values of the clipped positions;
e) derive updated motion vectors that lead to a minimum matching cost based on the compensation images;
f) generate the prediction image based on the updated motion vectors; and
g) decode the current picture based on the prediction image.