Skip to content

O
opencv_contrib
Commit 7a18e788 authored by Vladislav Samsonov's avatar Vladislav Samsonov
Browse files
Options

Post-review fixes

parent 54e746be
Hide whitespace changes
Inline Side-by-side
Showing with 149 additions and 252 deletions
modules/optflow/include/opencv2/optflow.hpp
View file @ 7a18e788
......@@ -193,8 +193,6 @@ CV_EXPORTS_W Ptr<DenseOpticalFlow> createOptFlow_Farneback();
//! Additional interface to the SparseToDenseFlow algorithm - calcOpticalFlowSparseToDense()
CV_EXPORTS_W Ptr<DenseOpticalFlow> createOptFlow_SparseToDense();
CV_EXPORTS_W Ptr<DenseOpticalFlow> createOptFlow_BlockMatching();
//! @}
} //optflow
......
modules/optflow/include/opencv2/optflow/pcaflow.hpp
View file @ 7a18e788
......@@ -8,7 +8,7 @@ copy or use the software.
For Open Source Computer Vision Library
(3-clause BSD License)
Copyright (C) 2013, OpenCV Foundation, all rights reserved.
Copyright (C) 2016, OpenCV Foundation, all rights reserved.
Third party copyrights are property of their respective owners.
Redistribution and use in source and binary forms, with or without modification,
......@@ -37,6 +37,19 @@ or tort (including negligence or otherwise) arising in any way out of
the use of this software, even if advised of the possibility of such damage.
*/
/*
Implementation of the PCAFlow algorithm from the following paper:
http://files.is.tue.mpg.de/black/papers/cvpr2015_pcaflow.pdf
@inproceedings{Wulff:CVPR:2015,
title = {Efficient Sparse-to-Dense Optical Flow Estimation using a Learned Basis and Layers},
author = {Wulff, Jonas and Black, Michael J.},
booktitle = { IEEE Conf. on Computer Vision and Pattern Recognition (CVPR) 2015},
month = jun,
year = {2015}
}
*/
#ifndef __OPENCV_OPTFLOW_PCAFLOW_HPP__
#define __OPENCV_OPTFLOW_PCAFLOW_HPP__
......@@ -47,10 +60,13 @@ namespace cv
{
namespace optflow
{
namespace pcaflow
{
class Prior
/*
* This class can be used for imposing a learned prior on the resulting optical flow.
* Solution will be regularized according to this prior.
* You need to generate appropriate prior file with "learn_prior.py" script beforehand.
*/
class PCAPrior
{
private:
Mat L1;
......@@ -59,7 +75,7 @@ private:
Mat c2;
public:
Prior( const char *pathToPrior );
PCAPrior( const char *pathToPrior );
int getPadding() const { return L1.size().height; }
......@@ -67,12 +83,11 @@ public:
void fillConstraints( float *A1, float *A2, float *b1, float *b2 ) const;
};
}
class OpticalFlowPCAFlow : public DenseOpticalFlow
{
protected:
const pcaflow::Prior *prior;
const PCAPrior *prior;
const Size basisSize;
const float sparseRate; // (0 .. 0.1)
const float retainedCornersFraction; // [0 .. 1]
......@@ -80,7 +95,7 @@ protected:
const float dampingFactor;
public:
OpticalFlowPCAFlow( const pcaflow::Prior *_prior = 0, const Size _basisSize = Size( 18, 14 ),
OpticalFlowPCAFlow( const PCAPrior *_prior = 0, const Size _basisSize = Size( 18, 14 ),
float _sparseRate = 0.02, float _retainedCornersFraction = 0.7,
float _occlusionsThreshold = 0.0003, float _dampingFactor = 0.00002 );
......
modules/optflow/samples/optical_flow_evaluation.cpp
View file @ 7a18e788
......@@ -12,7 +12,7 @@ using namespace optflow;
const String keys = "{help h usage ? | | print this message }"
"{@image1 | | image1 }"
"{@image2 | | image2 }"
"{@algorithm | | [farneback, simpleflow, tvl1, deepflow, pcaflow, blockmatching or sparsetodenseflow] }"
"{@algorithm | | [farneback, simpleflow, tvl1, deepflow, pcaflow or sparsetodenseflow] }"
"{@groundtruth | | path to the .flo file (optional), Middlebury format }"
"{m measure |endpoint| error measure - [endpoint or angular] }"
"{r region |all | region to compute stats about [all, discontinuities, untextured] }"
......@@ -258,8 +258,6 @@ int main( int argc, char** argv )
algorithm = createOptFlow_DeepFlow();
else if ( method == "sparsetodenseflow" )
algorithm = createOptFlow_SparseToDense();
else if ( method == "blockmatching" )
algorithm = createOptFlow_BlockMatching();
else if ( method == "pcaflow" )
algorithm = createOptFlow_PCAFlow();
else
......
modules/optflow/src/block_matching.cpp deleted 100644 → 0
View file @ 54e746be
/*M///////////////////////////////////////////////////////////////////////////////////////
//
// IMPORTANT: READ BEFORE DOWNLOADING, COPYING, INSTALLING OR USING.
//
// By downloading, copying, installing or using the software you agree to this license.
// If you do not agree to this license, do not download, install,
// copy or use the software.
//
//
// License Agreement
// For Open Source Computer Vision Library
//
// Copyright (C) 2000-2008, Intel Corporation, all rights reserved.
// Copyright (C) 2009, Willow Garage Inc., all rights reserved.
// Third party copyrights are property of their respective owners.
//
// Redistribution and use in source and binary forms, with or without modification,
// are permitted provided that the following conditions are met:
//
// * Redistribution's of source code must retain the above copyright notice,
// this list of conditions and the following disclaimer.
//
// * Redistribution's in binary form must reproduce the above copyright notice,
// this list of conditions and the following disclaimer in the documentation
// and/or other materials provided with the distribution.
//
// * The name of the copyright holders may not be used to endorse or promote products
// derived from this software without specific prior written permission.
//
// This software is provided by the copyright holders and contributors "as is" and
// any express or implied warranties, including, but not limited to, the implied
// warranties of merchantability and fitness for a particular purpose are disclaimed.
// In no event shall the Intel Corporation or contributors be liable for any direct,
// indirect, incidental, special, exemplary, or consequential damages
// (including, but not limited to, procurement of substitute goods or services;
// loss of use, data, or profits; or business interruption) however caused
// and on any theory of liability, whether in contract, strict liability,
// or tort (including negligence or otherwise) arising in any way out of
// the use of this software, even if advised of the possibility of such damage.
//
//M*/
#include "precomp.hpp"
namespace cv
{
namespace optflow
{
class OpticalFlowBlockMatching : public DenseOpticalFlow
{
protected:
int windowSize;
int blockSize;
inline float submatrixAbsDiff( int x0, int y0, const Mat &I0, int x1, int y1, const Mat &I1 ) const;
public:
OpticalFlowBlockMatching() : windowSize( 3 ), blockSize( 8 ){};
void calc( InputArray I0, InputArray I1, InputOutputArray flow );
void collectGarbage();
};
inline float OpticalFlowBlockMatching::submatrixAbsDiff( int x0, int y0, const Mat &I0, int x1, int y1,
const Mat &I1 ) const
{
float error = 0;
const Size size = I0.size();
for ( int i = -windowSize; i <= windowSize; ++i )
{
if ( i + y0 < 0 || i + y0 >= size.height || i + y1 < 0 || i + y1 >= size.height )
{
error += 1;
continue;
}
const Vec3f *I0X = I0.ptr<Vec3f>( i + y0 );
const Vec3f *I1X = I1.ptr<Vec3f>( i + y1 );
for ( int j = -windowSize; j <= windowSize; ++j )
{
if ( j + x0 < 0 || j + x0 >= size.width || j + x1 < 0 || j + x1 >= size.width )
{
error += 1;
continue;
}
const Vec3f diff = I0X[j + x0] - I1X[j + x1];
error += abs( diff[0] );
error += abs( diff[1] );
error += abs( diff[2] );
}
}
return error;
}
void OpticalFlowBlockMatching::calc( InputArray I0, InputArray I1, InputOutputArray flow_out )
{
CV_Assert( I0.channels() == 3 );
CV_Assert( I1.channels() == 3 );
Size size = I0.size();
CV_Assert( size == I1.size() );
flow_out.create( size, CV_32FC2 );
Mat flow = flow_out.getMat();
Mat from = I0.getMat();
Mat to = I1.getMat();
from.convertTo( from, CV_32FC3, 1.0 / 255.0 );
to.convertTo( to, CV_32FC3, 1.0 / 255.0 );
const float distNormalize = blockSize * sqrt( 2 );
for ( int y0 = 0; y0 < size.height; ++y0 )
{
Vec2f *flowX = flow.ptr<Vec2f>( y0 );
const int yEnd = std::min( size.height - 1, y0 + blockSize );
for ( int x0 = 0; x0 < size.width; ++x0 )
{
float minDiff = 1e10;
Vec2f du( 0, 0 );
const int xEnd = std::min( size.width - 1, x0 + blockSize );
for ( int y1 = std::max( 0, y0 - blockSize ); y1 <= yEnd; ++y1 )
for ( int x1 = std::max( 0, x0 - blockSize ); x1 <= xEnd; ++x1 )
{
const float distance = sqrt( ( x0 - x1 ) * ( x0 - x1 ) + ( y0 - y1 ) * ( y0 - y1 ) ) / distNormalize;
const float kernel = 1.0 + 0.5 * distance;
const float diff = kernel * submatrixAbsDiff( x0, y0, from, x1, y1, to );
if ( diff < minDiff )
{
minDiff = diff;
du = Vec2f( ( x1 - x0 ), ( y1 - y0 ) );
}
}
flowX[x0] = du;
}
}
}
void OpticalFlowBlockMatching::collectGarbage() {}
Ptr<DenseOpticalFlow> createOptFlow_BlockMatching() { return makePtr<OpticalFlowBlockMatching>(); }
}
}
modules/optflow/src/learn_prior.py
View file @ 7a18e788
import os, sys
#!/usr/bin/env python
import os
import sys
import numpy as np
import cv2
import struct
import argparse
from math import sqrt
basis_size = (14, 18)
lambd = 0.2
argparser = argparse.ArgumentParser(
description='''Use this script to generate prior for using with PCAFlow.
Basis size here must match corresponding parameter in the PCAFlow.
Gamma should be selected experimentally.''')
argparser.add_argument('-f',
'--files',
nargs='+',
help='List of optical flow .flo files for learning',
required=True)
argparser.add_argument('-o',
'--output',
help='Output file for prior',
required=True)
argparser.add_argument('--width',
type=int,
help='Size of the basis first dimension',
required=True,
default=18)
argparser.add_argument('--height',
type=int,
help='Size of the basis second dimension',
required=True,
default=14)
argparser.add_argument(
'-g',
'--gamma',
type=float,
help='Amount of regularization. The greater this parameter, the bigger will be an impact of the regularization.',
required=True)
args = argparser.parse_args()
basis_size = (args.height, args.width)
gamma = args.gamma
def find_flo(pp):
f = []
for p in pp:
if os.path.isfile(p):
f.append(p)
else:
for root, subdirs, files in os.walk(p):
f += map(lambda x: os.path.join(root, x), filter(lambda x: x.split('.')[-1] == 'flo', files))
return list(set(f))
f = []
for p in pp:
if os.path.isfile(p):
f.append(p)
else:
for root, subdirs, files in os.walk(p):
f += map(lambda x: os.path.join(root, x),
filter(lambda x: x.split('.')[-1] == 'flo', files))
return list(set(f))
def load_flo(flo):
with open(flo) as f:
magic = np.fromfile(f, np.float32, count=1)
if 202021.25 != magic:
print('Magic number incorrect. Invalid .flo file')
else:
w = np.fromfile(f, np.int32, count=1)
h = np.fromfile(f, np.int32, count=1)
print('Reading %dx%d flo file %s' % (w, h, flo))
data = np.fromfile(f, np.float32, count=2*w*h)
# Reshape data into 3D array (columns, rows, bands)
flow = np.resize(data, (h, w, 2))
return flow[:,:,0], flow[:,:,1]
magic = np.fromfile(f, np.float32, count=1)[0]
if 202021.25 != magic:
print('Magic number incorrect. Invalid .flo file')
else:
w = np.fromfile(f, np.int32, count=1)[0]
h = np.fromfile(f, np.int32, count=1)[0]
print('Reading %dx%d flo file %s' % (w, h, flo))
data = np.fromfile(f, np.float32, count=2 * w * h)
# Reshape data into 3D array (columns, rows, bands)
flow = np.reshape(data, (h, w, 2))
return flow[:, :, 0], flow[:, :, 1]
def get_w(m):
s = m.shape
w = cv2.dct(m)
w *= 2.0 / sqrt(s[0] * s[1])
#w[0,0] *= 0.5
w[:,0] *= sqrt(0.5)
w[0,:] *= sqrt(0.5)
w = w[0:basis_size[0],0:basis_size[1]].transpose().flatten()
return w
s = m.shape
w = cv2.dct(m)
w *= 2.0 / sqrt(s[0] * s[1])
#w[0,0] *= 0.5
w[:, 0] *= sqrt(0.5)
w[0, :] *= sqrt(0.5)
w = w[0:basis_size[0], 0:basis_size[1]].transpose().flatten()
return w
w1 = []
w2 = []
for flo in find_flo(sys.argv[1:]):
x,y = load_flo(flo)
w1.append(get_w(x))
w2.append(get_w(y))
for flo in find_flo(args.files):
x, y = load_flo(flo)
w1.append(get_w(x))
w2.append(get_w(y))
w1mean = sum(w1) / len(w1)
w2mean = sum(w2) / len(w2)
for i in xrange(len(w1)):
w1[i] -= w1mean
w1[i] -= w1mean
for i in xrange(len(w2)):
w2[i] -= w2mean
w2[i] -= w2mean
Q1 = sum([w1[i].reshape(-1,1).dot(w1[i].reshape(1,-1)) for i in xrange(len(w1))]) / len(w1)
Q2 = sum([w2[i].reshape(-1,1).dot(w2[i].reshape(1,-1)) for i in xrange(len(w2))]) / len(w2)
Q1 = sum([w1[i].reshape(-1, 1).dot(w1[i].reshape(1, -1))
for i in xrange(len(w1))]) / len(w1)
Q2 = sum([w2[i].reshape(-1, 1).dot(w2[i].reshape(1, -1))
for i in xrange(len(w2))]) / len(w2)
Q1 = np.matrix(Q1)
Q2 = np.matrix(Q2)
if len(w1) > 1:
while True:
try:
L1 = np.linalg.cholesky(Q1)
break
except np.linalg.linalg.LinAlgError:
mev = min(np.linalg.eig(Q1)[0]).real
assert(mev < 0)
print('Q1', mev)
if -mev < 1e-6:
mev = -1e-6
Q1 += (-mev*1.000001) * np.identity(Q1.shape[0])
while True:
try:
L2 = np.linalg.cholesky(Q2)
break
except np.linalg.linalg.LinAlgError:
mev = min(np.linalg.eig(Q2)[0]).real
assert(mev < 0)
print('Q2', mev)
if -mev < 1e-6:
mev = -1e-6
Q2 += (-mev*1.000001) * np.identity(Q2.shape[0])
while True:
try:
L1 = np.linalg.cholesky(Q1)
break
except np.linalg.linalg.LinAlgError:
mev = min(np.linalg.eig(Q1)[0]).real
assert (mev < 0)
print('Q1', mev)
if -mev < 1e-6:
mev = -1e-6
Q1 += (-mev * 1.000001) * np.identity(Q1.shape[0])
while True:
try:
L2 = np.linalg.cholesky(Q2)
break
except np.linalg.linalg.LinAlgError:
mev = min(np.linalg.eig(Q2)[0]).real
assert (mev < 0)
print('Q2', mev)
if -mev < 1e-6:
mev = -1e-6
Q2 += (-mev * 1.000001) * np.identity(Q2.shape[0])
else:
L1 = np.identity(Q1.shape[0])
L2 = np.identity(Q2.shape[0])
L1 = np.identity(Q1.shape[0])
L2 = np.identity(Q2.shape[0])
L1 = np.linalg.inv(L1) * lambd
L2 = np.linalg.inv(L2) * lambd
L1 = np.linalg.inv(L1) * gamma
L2 = np.linalg.inv(L2) * gamma
assert(L1.shape == L2.shape)
assert(L1.shape[0] == L1.shape[1])
assert (L1.shape == L2.shape)
assert (L1.shape[0] == L1.shape[1])
f = open('cov.dat', 'wb')
f = open(args.output, 'wb')
f.write(struct.pack('I', L1.shape[0]))
f.write(struct.pack('I', L1.shape[1]))
for i in xrange(L1.shape[0]):
for j in xrange(L1.shape[1]):
f.write(struct.pack('f', L1[i,j]))
for j in xrange(L1.shape[1]):
f.write(struct.pack('f', L1[i, j]))
for i in xrange(L2.shape[0]):
for j in xrange(L2.shape[1]):
f.write(struct.pack('f', L2[i,j]))
for j in xrange(L2.shape[1]):
f.write(struct.pack('f', L2[i, j]))
b1 = L1.dot(w1mean.reshape(-1,1))
b2 = L2.dot(w2mean.reshape(-1,1))
b1 = L1.dot(w1mean.reshape(-1, 1))
b2 = L2.dot(w2mean.reshape(-1, 1))
assert(L1.shape[0] == b1.shape[0])
assert (L1.shape[0] == b1.shape[0])
for i in xrange(b1.shape[0]):
f.write(struct.pack('f', b1[i,0]))
f.write(struct.pack('f', b1[i, 0]))
for i in xrange(b2.shape[0]):
f.write(struct.pack('f', b2[i,0]))
f.write(struct.pack('f', b2[i, 0]))
f.close()
\ No newline at end of file
f.close()
modules/optflow/src/pcaflow.cpp
View file @ 7a18e788
......@@ -48,7 +48,7 @@ namespace cv
namespace optflow
{
OpticalFlowPCAFlow::OpticalFlowPCAFlow( const pcaflow::Prior *_prior, const Size _basisSize, float _sparseRate,
OpticalFlowPCAFlow::OpticalFlowPCAFlow( const PCAPrior *_prior, const Size _basisSize, float _sparseRate,
float _retainedCornersFraction, float _occlusionsThreshold,
float _dampingFactor )
: prior( _prior ), basisSize( _basisSize ), sparseRate( _sparseRate ),
......@@ -60,15 +60,6 @@ OpticalFlowPCAFlow::OpticalFlowPCAFlow( const pcaflow::Prior *_prior, const Size
CV_Assert( occlusionsThreshold > 0 );
}
inline float eDistSq( const Point2f &p1, const Point2f &p2 )
{
const float dx = p1.x - p2.x;
const float dy = p1.y - p2.y;
return dx * dx + dy * dy;
}
inline float eNormSq( const Point2f &v ) { return v.x * v.x + v.y * v.y; }
template <typename T> static inline int mathSign( T val ) { return ( T( 0 ) < val ) - ( val < T( 0 ) ); }
static inline void symOrtho( double a, double b, double &c, double &s, double &r )
......@@ -224,7 +215,7 @@ void OpticalFlowPCAFlow::removeOcclusions( UMat &from, UMat &to, std::vector<Poi
if ( predictedStatus[i] )
{
Point2f flowDiff = features[i] - backwardFeatures[i];
if ( eNormSq( flowDiff ) <= threshold )
if ( flowDiff.dot( flowDiff ) <= threshold )
{
features[j] = features[i];
predictedFeatures[j] = predictedFeatures[i];
......@@ -462,10 +453,7 @@ void OpticalFlowPCAFlow::collectGarbage() {}
Ptr<DenseOpticalFlow> createOptFlow_PCAFlow() { return makePtr<OpticalFlowPCAFlow>(); }
namespace pcaflow
{
Prior::Prior( const char *pathToPrior )
PCAPrior::PCAPrior( const char *pathToPrior )
{
FILE *f = fopen( pathToPrior, "r" );
CV_Assert( f );
......@@ -487,7 +475,7 @@ Prior::Prior( const char *pathToPrior )
fclose( f );
}
void Prior::fillConstraints( float *A1, float *A2, float *b1, float *b2 ) const
void PCAPrior::fillConstraints( float *A1, float *A2, float *b1, float *b2 ) const
{
memcpy( A1, L1.ptr<float>(), L1.size().area() * sizeof( float ) );
memcpy( A2, L2.ptr<float>(), L2.size().area() * sizeof( float ) );
......@@ -496,4 +484,3 @@ void Prior::fillConstraints( float *A1, float *A2, float *b1, float *b2 ) const
}
}
}
}
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment