Motion vector regions
Coding techniques for a video image compression system involve improving an image quality of a sequence of two or more bi-directionally predicted intermediate frames, where each of the frames includes multiple pixels. One method involves determining a brightness value of at least one pixel of each bi-directionally predicted intermediate frame in the sequence as an equal average of brightness values of pixels in non-bidirectionally predicted frames bracketing the sequence of bi-directionally predicted intermediate frames. The brightness values of the pixels in at least one of the non-bidirectionally predicted frames is converted from a non-linear representation.
1. A computer-implemented method for video imaging comprising:
determining a first motion vector to a single reference frame for a first plurality of 4×4 regions corresponding to a first motion vector region of a macroblock;
determining a second motion vector to the single reference frame for a second plurality of 4×4 regions corresponding to a second motion vector region of the macroblock, wherein the second motion vector is determined using a motion vector delta; and
performing motion compensation based on the first motion vector and the second motion vector to determine pixel luminance values for the macroblock including at least the first and second motion vector regions, the first and second motion vector regions each being signaled in a received bitstream, wherein boundaries of the first and second motion vector regions correspond to one or more boundaries of the corresponding plurality of 4×4 regions.
2. The method of claim 1 wherein the single reference frame is selected from three frames, the three frames includes a most recent previous interpredicted frame in decoding order and a previous intraframe in decoding order.
3. The method of claim 1 further comprising outputting the determined pixel luminance values of the macroblock.
4. The method of claim 1 wherein the macroblock is sized 16×16.
5. The method of claim 4 wherein the first motion vector region is sized 8×16 and includes four 4×4 regions.
6. A computer-implemented method for video imaging comprising:
determining a first motion vector to a single reference frame for a first motion vector region, the first motion vector region being sized 8×16 or 16×8 and corresponding to at least four 4×4 regions;
determining, using a motion vector delta, a second motion vector to the single reference frame for a second motion vector region;
performing motion compensation to determine pixel luminance values for a 16×16 macroblock including at least the first and second motion vector regions, wherein the correspondence between the first motion vector region and the at least four 4×4 regions is signaled in a received bitstream;
wherein the single reference frame is selected from three frames, the three frames including a most recent previous interpredicted frame in decoding order and a previous intraframe in decoding order.
7. The method of claim 6 wherein the single reference frame is selected using 2 bits in the received bitstream.