dirac::BlockMatcher Class Reference

Class to do block matching. More...

#include <block_match.h>

List of all members.

Public Member Functions
	BlockMatcher (const PicArray &ref_data, const PicArray &pic_data, const OLBParams &bparams, const MvArray &mv_array, const TwoDArray< MvCostData > &cost_array)
	Constructor.
void	FindBestMatch (int xpos, int ypos, const CandidateList &cand_list, const MVector &mv_prediction, float lambda)
	Find the best matching vector from a list of candidates.
void	FindBestMatchSubp (int xpos, int ypos, const CandidateList &cand_list, const MVector &mv_prediction, float lambda)
	Find the best matching vector from a list of candidates, to sub-pixel accuracy (TBC: merge with FindBestMatch).

---

Detailed Description

Class to do block matching.

Definition at line 104 of file block_match.h.

---

Constructor & Destructor Documentation

dirac::BlockMatcher::BlockMatcher (  const PicArray &  ref_data, const PicArray &  pic_data, const OLBParams &  bparams, const MvArray &  mv_array, const TwoDArray< MvCostData > &  cost_array )

Constructor. Constructor Parameters: ref_data  the reference picture component pic_data  the picture being matched bparams  the (overlapped) block parameters to be used for the matching mv_array  the array of vectors we're going to write into cost_array  the array of costs we're going to write into Definition at line 217 of file block_match.cpp.

---

Member Function Documentation

void dirac::BlockMatcher::FindBestMatch (  int  xpos, int  ypos, const CandidateList &  cand_list, const MVector &  mv_prediction, float  lambda )

Find the best matching vector from a list of candidates. Find the best matching vector from a list of candidates. Parameters: xpos  the horizontal location of the block being matched ypos  the vertical location of the block being matched cand_list  the list of candidate vectors mv_prediction  the prediction for the motion vector lambda  the Lagrangian parameter Definition at line 230 of file block_match.cpp. References dirac::BlockDiff::Diff(), dirac::GetVar(), dirac::TwoDArray< T >::LengthX(), dirac::TwoDArray< T >::LengthY(), dirac::MvCostData::mvcost, dirac::MvCostData::SAD, dirac::BlockDiffParams::SetBlockLimits(), dirac::MvCostData::SetTotal(), dirac::MvCostData::total, dirac::MotionVector< T >::x, dirac::OLBParams::Xblen(), dirac::BlockDiffParams::Xl(), dirac::BlockDiffParams::Xp(), dirac::MotionVector< T >::y, dirac::OLBParams::Yblen(), dirac::BlockDiffParams::Yl(), and dirac::BlockDiffParams::Yp().

void dirac::BlockMatcher::FindBestMatchSubp (  int  xpos, int  ypos, const CandidateList &  cand_list, const MVector &  mv_prediction, float  lambda )

Find the best matching vector from a list of candidates, to sub-pixel accuracy (TBC: merge with FindBestMatch). Find the best matching vector from a list of candidates. Parameters: xpos  the horizontal location of the block being matched ypos  the vertical location of the block being matched cand_list  the list of candidate vectors mv_prediction  the prediction for the motion vector lambda  the Lagrangian parameter Definition at line 354 of file block_match.cpp. References dirac::BlockDiff::Diff(), dirac::GetVar(), dirac::TwoDArray< T >::LengthX(), dirac::TwoDArray< T >::LengthY(), dirac::MvCostData::mvcost, dirac::MvCostData::SAD, dirac::BlockDiffParams::SetBlockLimits(), dirac::MvCostData::SetTotal(), dirac::MvCostData::total, dirac::MotionVector< T >::x, dirac::BlockDiffParams::Xl(), dirac::BlockDiffParams::Xp(), dirac::MotionVector< T >::y, dirac::BlockDiffParams::Yl(), and dirac::BlockDiffParams::Yp().

---

The documentation for this class was generated from the following files:

• C:/guliverkli/src/filters/parser/diracsplitter/libdirac/libdirac_motionest/block_match.h
• C:/guliverkli/src/filters/parser/diracsplitter/libdirac/libdirac_motionest/block_match.cpp

---