A class for doing differences with half-pixel accurate vectors.
#include <me_utils.h>
|
| BlockDiffHalfPel (const PicArray &ref, const PicArray &pic) |
| Constructor, initialising the reference and picture data.
|
| ~BlockDiffHalfPel () |
| Destructor.
|
float | Diff (const BlockDiffParams &dparams, const MVector &mv) |
| Do the difference, returning SAD.
|
void | Diff (const BlockDiffParams &dparams, const MVector &mv, const float mvcost, const float lambda, MvCostData &best_costs, MVector &best_mv) |
| Do the actual difference, overwriting the best MV so far if appropriate.
|
| BlockDiffUp (const PicArray &ref, const PicArray &pic) |
| Constructor, initialising the reference and picture data.
|
virtual | ~BlockDiffUp () |
| Destructor.
|
| BlockDiff (const PicArray &ref, const PicArray &pic) |
| Constructor, initialising the reference and picture data.
|
virtual | ~BlockDiff () |
| Destructor.
|
◆ BlockDiffHalfPel() [1/2]
dirac::BlockDiffHalfPel::BlockDiffHalfPel |
( |
const PicArray & | ref, |
|
|
const PicArray & | pic ) |
◆ ~BlockDiffHalfPel()
dirac::BlockDiffHalfPel::~BlockDiffHalfPel |
( |
| ) |
|
|
inline |
◆ BlockDiffHalfPel() [2/2]
◆ Diff() [1/2]
Do the difference, returning SAD
- Parameters
-
dparams | block parameters |
mv | the motion vector being used |
Implements dirac::BlockDiffUp.
◆ Diff() [2/2]
Do the actual difference, overwriting the best MV so far if appropriate, and bailing out if we do worse
- Parameters
-
dparams | block parameters |
mv | the motion vector being used |
mvcost | the (prediction) cost of the motion vector mv |
lambda | the weighting to be given to mvcost |
best_costs | the best Lagrangian costs obtained yet |
best_mv | the MV giving the best Lagrangian costs so far |
Implements dirac::BlockDiffUp.
◆ operator=()
The documentation for this class was generated from the following file:
© 2004 British Broadcasting Corporation.
Dirac code licensed under the Mozilla Public License (MPL) Version 1.1.
HTML documentation generated by Dimitri van Heesch's
excellent Doxygen tool.