Co-located motion vector storage
A method for processing a video bitstream is disclosed. The method generally includes the steps of (A) determining co-located motion information for a plurality of co-located sub-blocks in a co-located macroblock of the video bitstream, (B) storing the co-located motion information for (i) at least a first three of the co-located sub-blocks along a left side of the co-located macroblock, (ii) at least a second three of the co-located sub-blocks along a right side of the co-located macroblock and (iii) less than all of the co-located sub-blocks and (C) inferring current motion information for a current macroblock co-located in a different picture from the co-located macroblock using the co-located motion information that was stored.
1. A method for processing a video bitstream, comprising the steps of:
(A) determining co-located motion information of a plurality of co-located sub-blocks in a co-located macroblock of said video bitstream;
(B) storing said co-located motion information of (i) at least a first three of said co-located sub-blocks along a left side of said co-located macroblock, (ii) at least a second three of said co-located sub-blocks along a right side of said co-located macroblock and (iii) less than all of said co-located sub-blocks;
(C) inferring current motion information of a current macroblock co-located in a different picture from said co-located macroblock using said co-located motion information that was stored, wherein said current motion information is inferred without measuring an inter-picture difference; and
(D) generating a processed signal by one of (i) compressing said video bitstream using said current motion information and (ii) decompressing said video bitstream using said current motion information.
2. The method according to claim 1 , wherein (i) said first three co-located sub-blocks comprise a sub-block number 0 , a sub-block number 8 and a sub-block number 10 and (ii) said second three co-located sub-blocks comprise a sub-block number 5 , a sub-block number 13 and a sub-block number 15 per an H.264 numbering.
3. The method according to claim 1 , wherein step (B) comprises the sub-steps of:
storing said co-located motion information of at least a first four of said co-located sub-blocks along said left side; and
storing said co-located motion information of at least a second four of said co-located sub-blocks along said right side.
4. The method according to claim 3 , wherein said first four co-located sub-blocks comprise a sub-block number 0 , a sub-block number 2 , a sub-block number 8 and a sub-block number 10 of said co-located macroblock per an H.264 numbering.
5. The method according to claim 3 , wherein said second four co-located sub-blocks comprise a sub-block number 5 , a sub-block number 7 , a sub-block number 13 and a sub-block number 15 of said co-located macroblock per an H.264 numbering.
6. The method according to claim 1 , wherein step (C) comprises the sub-steps of:
inferring a first portion of said current motion information of a current sub-block number 10 of an even-macroblock row frame macroblock from a first portion of said co-located motion information of a co-located sub-block number 2 of said co-located macroblock per an H.264 numbering; and
inferring a second portion of said current motion information of a current sub-block number 0 of an odd-macroblock row frame macroblock in said current macroblock from a second portion of said co-located motion information of a co-located sub-block number 8 of said co-located macroblock per said H.264 numbering.
7. The method according to claim 1 , wherein step (C) comprises the sub-steps of:
inferring a first portion of said current motion information of a current sub-block number 15 of an even-macroblock row macroblock from a first portion of said co-located motion information of a co-located sub-block number 7 of said co-located macroblock per an H.264 numbering; and
inferring a second portion of said current motion information of a current sub-block number 5 of an odd-macroblock row frame macroblock in said current macroblock from a second portion of said co-located motion information of a co-located sub-block number 13 of said co-located macroblock per said H.264 numbering.
8. A method for processing a video bitstream, comprising the steps of:
(A) determining co-located motion information of a plurality of co-located sub-blocks in a co-located macroblock of said video, said co-located macroblock comprising a frame macroblock in an adaptive field/frame picture while in a first mode and a field macroblock while in a second mode;
(B) storing said co-located motion information of between at least six and less than all of said co-located sub-blocks;
(C) inferring current motion information of a current macroblock co-located in a different picture from said co-located macroblock using said co-located motion information that was stored, wherein said current motion information is inferred without measuring an inter-picture difference; and
(D) generating a processed signal by one of (i) compressing said video bitstream using said current motion information and (ii) decompressing said video bitstream using said current motion information.
9. The method according to claim 8 , wherein said co-located macroblock comprises said field macroblock.
10. The method according to claim 9 , wherein step (B) comprises the sub-step of:
storing said co-located motion information of not more than eight of said co-located sub-blocks.
11. The method according to claim 8 , wherein said co-located macroblock comprises two frame macroblocks.
12. The method according to claim 11 , wherein step (B) comprises the sub-step of:
storing said co-located motion information of not more than ten of said co-located sub-blocks between said two frame macroblocks.
13. An apparatus comprising:
a memory; and
a circuit configured to (i) determine co-located motion information of a plurality of co-located sub-blocks in a co-located macroblock of a video bitstream, (ii) store said co-located motion information of (a) at least three of said co-located sub-blocks along a left side of said co-located macroblock, (b) at least three of said co-located sub-blocks along a right side of said co-located macroblock and (c) less than all of said co-located sub-blocks in said memory and (iii) infer current motion information of a current macroblock co-located in a different picture from said co-located macroblock using said co-located motion information stored in said memory, wherein said current motion information is inferred without measuring an inter-picture difference.
14. The apparatus according to claim 13 , wherein said co-located macroblock comprises two field macroblocks.
15. The apparatus according to claim 14 , wherein said current macroblock comprises two frame macroblocks arranged vertically.
16. The apparatus according to claim 13 , wherein said current motion information of a first half of said current macroblock is inferred only from said co-located motion information of a first frame macroblock of said co-located macroblock.
17. The apparatus according to claim 16 , wherein said current motion information of a second half of said current macroblock is inferred only from said co-located motion information of a second frame macroblock of said co-located macroblock.
18. An apparatus comprising:
a memory; and
a circuit configured to (i) determine co-located motion information of a plurality of co-located sub-blocks in a co-located macroblock of a video bitstream, said co-located macroblock comprising a frame macroblock in an adaptive field/frame picture while in a first mode and a field macroblock while in a second mode, (ii) store said co-located motion information of between at least six and less than all of said co-located sub-blocks in said memory and (iii) infer a plurality of current motion information of a current macroblock co-located in a different picture from said co-located macroblock using said co-located motion information stored in said memory, wherein said current motion information is inferred without measuring an inter-picture difference.
19. The apparatus according to claim 18 , wherein said co-located macroblock comprises said field macroblock.
20. The apparatus according to claim 19 , wherein said circuit is further configured to use said co-location motion information from only four of said co-located sub-blocks.
21. The apparatus according to claim 18 , wherein said co-located macroblock comprises two frame macroblocks arranged vertically.
22. The apparatus according to claim 21 , wherein said circuit is further configured to infer said current motion information of each quadrant of said current macroblock from said co-located motion information in a corresponding quadrant of said co-located macroblock.
23. A method for processing a video bitstream, comprising the steps of:
(A) determining co-located motion information for a plurality of co-located sub-blocks in a co-located macroblock of said video bitstream;
(B) storing said co-located motion information for (i) at least a first three of said co-located sub-blocks along a left side of said co-located macroblock, (ii) at least a second three of said co-located sub-blocks along a right side of said co-located macroblock and (iii) less than all of said co-located sub-blocks;
(C) inferring current motion information for a current macroblock co-located in a different picture from said co-located macroblock using said co-located motion information that was stored; and
(D) generating a processed signal by one of (i) compressing said video bitstream using said current motion information and (ii) decompressing said video bitstream using said current motion information, wherein:
(i) step (B) comprises the sub-steps of
storing said co-located motion information of at least a first four of said co-located sub-blocks along said left side, and
storing said co-located motion information of at least a second four of said co-located sub-blocks along said right side;
(ii) said first four co-located sub-blocks comprise a sub-block number 0 , a sub-block number 2 , a sub-block number 8 and a sub-block number 10 of said co-located macroblock per an H.264 standard for numbering; and
(iii) step (C) comprises the sub-steps of:
inferring a first portion of said current motion information for a sub-block number 0 of an even-macroblock row frame macroblock in said current macroblock from a first portion of said co-located motion information for said sub-block number 0 of said co-located macroblock,
inferring a second portion of said current motion information for a sub-block number 10 of said even-macroblock row frame macroblock from a second portion of said co-located motion information for said sub-block number 2 of said co-located macroblock,
inferring a third portion of said current motion information for a sub-block number 0 of an odd-macroblock row frame macroblock in said current macroblock from a third portion of said co-located motion information for said sub-block number 8 of said co-located macroblock, and
inferring a fourth portion of said current motion information for a sub-block number 10 of said odd-macroblock row frame macroblock from a fourth portion of said co-located motion information for said sub-block 10 of said co-located macroblock.
24. A method for processing a video bitstream, comprising the steps of:
(A) determining co-located motion information for a plurality of co-located sub-blocks in a co-located macroblock of said video bitstream;
(B) storing said co-located motion information for (i) at least a first three of said co-located sub-blocks along a left side of said co-located macroblock, (ii) at least a second three of said co-located sub-blocks along a right side of said co-located macroblock and (iii) less than all of said co-located sub-blocks; and
(C) inferring current motion information for a current macroblock co-located in a different picture from said co-located macroblock using said co-located motion information that was stored; and
(D) generating a processed signal by one of (i) compressing said video bitstream using said current motion information and (ii) decompressing said video bitstream using said current motion information, wherein:
(i) step (B) comprises the sub-steps of
storing said co-located motion information of at least a first four of said co-located sub-blocks along said left side, and
storing said co-located motion information of at least a second four of said co-located sub-blocks along said right side;
(ii) said second four co-located sub-blocks comprise a sub-block number 5 , a sub-block number 7 , a sub-block number 13 and a sub-block number 15 of said co-located macroblock per an H.264 standard for numbering; and
(iii) step (C) comprises the sub-steps of
inferring a first portion of said current motion information for a sub-block number 5 of an even-macroblock row frame macroblock in said current macroblock from a first portion of said co-located motion information for said sub-block number 5 of said co-located macroblock,
inferring a second portion of said current motion information for a sub-block number 15 of said even-macroblock row macroblock from a second portion of said co-located motion information for said sub-block number 7 of said co-located macroblock,
inferring a third portion of said current motion information for a sub-block number 5 of an odd-macroblock row frame macroblock in said current macroblock from a third portion of said co-located motion information for said sub-block number 13 of said co-located macroblock, and
inferring a fourth portion of said current motion information for a sub-block number 15 of said odd-macroblock row frame macroblock from a fourth portion of said co-located motion information for said sub-block 15 of said co-located macroblock.