Method for computing position of integer grid reference sample for block level boundary sample gradient computation in bi-predictive optical flow computation and bi-predictive correction
A prediction method and reference sample technology, applied in the field of image processing, can solve the problem of video size becoming a problem
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0345] Let the upper-left sample of the current sub-block of the M×N size sample undergoing BPOF / BDOF bidirectional prediction be located at (sb_start_x, sb_start_y). Let (mvx0, mvy0) be the 1 / 16 pixel precision MV of the current subblock relative to the first reference frame L0, let (mvx1, mvxy1) be the 1 / 16 pixel precision MV of the current subblock relative to the second reference frame L1.
[0346] The 1 / 16 pixel precision position in L0 corresponding to the upper left position of the current sub-block in L0 is obtained by ((sb_start_x<<4)+mvx0, (sb_start_y<<4)+mvy0). The 1 / 16 pixel precision position in L1 corresponding to the upper left position of the current sub-block in L0 is obtained by ((sb_start_x<<4)+mvx1, (sb_start_y<<4)+mvy1). It will be appreciated that different sub-pixel precisions may be used, such as half-pixel, quarter-pixel or 1 / 8-pixel. The top-left position of the current sub-block is determined by using a corresponding left shift of 1, 2 or 3, respect...
Embodiment 2
[0359] In this embodiment, when BPOF / BDOF is used and the specific coding unit to which the current sub-block belongs does not use motion vector refinement (motion vector refinement, MVR), or the specific coding unit to which the current sub-block belongs uses both MVR and BDOF, but When the MVR does not generate an integer displacement relative to the initial MV pair during the integer distance correction stage of the MVR, the extended sample position value is derived through the procedure described in Embodiment 1. In other words, when both MVR and BDOF are used, the integer distance correction stage of MVR generates initial MV pairs, ie with zero displacement, using the procedure described in Example 1 to derive extended sample position values.
[0360] When the specific coding unit to which the current subblock belongs uses both MVR and BDOF, but MVR generates integer displacements relative to the original MV pair during the integer distance correction stage of MVR, that is...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


