opencv/opencv_contrib
1
0
Fork 0
mirror of https://github.com/opencv/opencv_contrib.git synced 2025-10-17 07:04:18 +08:00
Code Issues Releases Wiki Activity

Add methods for get/set weights of variational refiment part of DISFlow method

Browse Source
This commit is contained in:
vbystricky
2016-07-15 11:25:46 +03:00
parent 55e119ceec
commit 5a1fd4c873
2 changed files with 36 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

19
modules/optflow/include/opencv2/optflow.hpp
View File

@@ -302,6 +302,25 @@ public:
/** @copybrief getGradientDescentIterations @see getGradientDescentIterations */
CV_WRAP virtual void setVariationalRefinementIterations(int val) = 0;
/** @brief Weight of the smoothness term
@see setVariationalRefinementAlpha */
CV_WRAP virtual float getVariationalRefinementAlpha() const = 0;
/** @copybrief getVariationalRefinementAlpha @see getVariationalRefinementAlpha */
CV_WRAP virtual void setVariationalRefinementAlpha(float val) = 0;
/** @brief Weight of the color constancy term
@see setVariationalRefinementDelta */
CV_WRAP virtual float getVariationalRefinementDelta() const = 0;
/** @copybrief getVariationalRefinementDelta @see getVariationalRefinementDelta */
CV_WRAP virtual void setVariationalRefinementDelta(float val) = 0;
/** @brief Weight of the gradient constancy term
@see setVariationalRefinementGamma */
CV_WRAP virtual float getVariationalRefinementGamma() const = 0;
/** @copybrief getVariationalRefinementGamma @see getVariationalRefinementGamma */
CV_WRAP virtual void setVariationalRefinementGamma(float val) = 0;
/** @brief Whether to use mean-normalization of patches when computing patch distance. It is turned on
by default as it typically provides a noticeable quality boost because of increased robustness to
illumanition variations. Turn it off if you are certain that your sequence does't contain any changes

20
modules/optflow/src/dis_flow.cpp
View File

@@ -65,6 +65,9 @@ class DISOpticalFlowImpl : public DISOpticalFlow
int patch_stride;
int grad_descent_iter;
int variational_refinement_iter;
float variational_refinement_alpha;
float variational_refinement_gamma;
float variational_refinement_delta;
bool use_mean_normalization;
bool use_spatial_propagation;
@@ -84,6 +87,13 @@ class DISOpticalFlowImpl : public DISOpticalFlow
void setGradientDescentIterations(int val) { grad_descent_iter = val; }
int getVariationalRefinementIterations() const { return variational_refinement_iter; }
void setVariationalRefinementIterations(int val) { variational_refinement_iter = val; }
float getVariationalRefinementAlpha() const { return variational_refinement_alpha; }
void setVariationalRefinementAlpha(float val) { variational_refinement_alpha = val; }
float getVariationalRefinementDelta() const { return variational_refinement_delta; }
void setVariationalRefinementDelta(float val) { variational_refinement_delta = val; }
float getVariationalRefinementGamma() const { return variational_refinement_gamma; }
void setVariationalRefinementGamma(float val) { variational_refinement_gamma = val; }
bool getUseMeanNormalization() const { return use_mean_normalization; }
void setUseMeanNormalization(bool val) { use_mean_normalization = val; }
bool getUseSpatialPropagation() const { return use_spatial_propagation; }
@@ -161,6 +171,10 @@ DISOpticalFlowImpl::DISOpticalFlowImpl()
patch_stride = 4;
grad_descent_iter = 16;
variational_refinement_iter = 5;
variational_refinement_alpha = 20.f;
variational_refinement_gamma = 10.f;
variational_refinement_delta = 5.f;
border_size = 16;
use_mean_normalization = true;
use_spatial_propagation = true;
@@ -234,9 +248,9 @@ void DISOpticalFlowImpl::prepareBuffers(Mat &I0, Mat &I1)
spatialGradient(I0s[i], I0xs[i], I0ys[i]);
Ux[i].create(cur_rows, cur_cols);
Uy[i].create(cur_rows, cur_cols);
variational_refinement_processors[i]->setAlpha(20.0f);
variational_refinement_processors[i]->setDelta(5.0f);
variational_refinement_processors[i]->setGamma(10.0f);
variational_refinement_processors[i]->setAlpha(variational_refinement_alpha);
variational_refinement_processors[i]->setDelta(variational_refinement_delta);
variational_refinement_processors[i]->setGamma(variational_refinement_gamma);
variational_refinement_processors[i]->setSorIterations(5);
variational_refinement_processors[i]->setFixedPointIterations(variational_refinement_iter);
}