Unequal weight planar motion vector derivation
A system and method of planar motion vector derivation which, in some embodiments can employ an unequal weighted combination of adjacent motion vectors. In some embodiments, motion vector information associated with a bottom right pixel or block adjacent to a current coding unit can be derived from motion information associated with a top row or top neighboring row of a current coding unit and motion information associated with a left column or left neighboring column of a current coding unit. Weighted or non-weighted combinations of such values can be combined in a planar mode prediction model to derive associated motion information for bottom and/or right adjacent pixels or blocks.
1. A method of video coding comprising:
identifying a coding unit of a current frame having a top neighboring row, a left neighboring column, a bottom neighboring row and a right neighboring column, each in the current frame;
determining motion information associated with a bottom right neighboring pixel positioned at the intersection of said bottom neighboring row and said right neighboring column using motion information associated with said top neighboring row of the current frame and said left neighboring column of the current frame;
determining motion information associated with said right neighboring column, based at least in part upon said motion information associated with said bottom right neighboring pixel; and
encoding said coding unit.
2. The method of video coding of claim 1 further comprising:
determining motion information associated with said bottom neighboring row, based at least in part upon said motion information associated with said bottom right neighboring pixel.
3. The method of video coding of claim 2 wherein a planar coding mode is employed.
4. The method of video coding of claim 3 wherein said current coding unit is encoded according to HEVC.
5. The method of video coding of claim 3 wherein said current coding using is encoded according to VVC.
6. The method of video coding of claim 3 wherein said current coding using is encoded according to JVET.
7. The method of video coding of claim 2 further comprising:
determining a first weight value associated with said top neighboring row; and
determining a second weight value associated with said left neighboring column,
wherein said step of determining motion information associated with said bottom right neighboring pixel is based at least in part on a combination of said first weight value and said motion information associated with said top neighboring row with a combination of said second weight value with said motion information associated with said left neighboring column.
8. The method of video coding of claim 7 wherein a planar coding mode is employed.
9. The method of video coding of claim 8 wherein said current coding unit is encoded according to HEVC.
10. The method of video coding of claim 8 wherein said current coding using is encoded according to VVC.
11. The method of video coding of claim 8 wherein said current coding using is encoded according to JVET.
12. A system of video coding comprising:
storing in memory a coding unit of a current frame having a top neighboring row, a left neighboring column, a bottom neighboring row and a right neighboring column, each in the current frame;
determining and storing in said memory motion information associated with a bottom right neighboring pixel positioned at the intersection of said bottom neighboring row and said right neighboring column using motion information associated with said top neighboring row of the current frame and said left neighboring column of the current frame;
determining and storing in memory motion information associated with said right neighboring column, based at least in part upon said motion information associated with said bottom right neighboring pixel; and
encoding said coding unit.
13. The system of video coding of claim 12 further comprising:
determining and storing in memory motion information associated with said bottom neighboring row, based at least in part upon said motion information associated with said bottom right neighboring pixel.
14. The system of video coding of claim 13 wherein a planar coding mode is employed.
15. The system of video coding of claim 14 , wherein said current coding unit is encoded according to at least one of HEVC, JVET and VVC.
16. The system of video coding of claim 13 further comprising:
determining and storing in memory a first weight value associated with said top neighboring row; and
determining and storing in memory a second weight value associated with said left neighboring column;
wherein said step of determining motion information associated with said bottom right neighboring pixel is based at least in part on a combination of said first weight value and said motion information associated with said top neighboring row with a combination of said second weight value with said motion information associated with said left neighboring column.
17. The system of video coding of claim 16 wherein a planar coding mode is employed.
18. The system of video coding of claim 17 wherein said current coding unit is encoded according to HEVC.
19. The system of video coding of claim 17 wherein said current coding using is encoded according to VVC.
20. The system of video coding of claim 17 wherein said current coding using is encoded according to JVET.