From: Vladislav Vinogradov Date: Mon, 25 Feb 2013 10:33:00 +0000 (+0400) Subject: Super Resolution module X-Git-Tag: accepted/tizen/ivi/20140515.103456~1^2~903^2 X-Git-Url: http://review.tizen.org/git/?a=commitdiff_plain;h=7a0d6f773374969eb3df1ff814e9d865b8f5ece8;p=profile%2Fivi%2Fopencv.git Super Resolution module --- diff --git a/modules/java/generator/rst_parser.py b/modules/java/generator/rst_parser.py index a9f5e04..33dae44 100755 --- a/modules/java/generator/rst_parser.py +++ b/modules/java/generator/rst_parser.py @@ -1,7 +1,7 @@ #/usr/bin/env python import os, sys, re, string, fnmatch -allmodules = ["core", "flann", "imgproc", "ml", "highgui", "video", "features2d", "calib3d", "objdetect", "legacy", "contrib", "gpu", "androidcamera", "java", "python", "stitching", "ts", "photo", "nonfree", "videostab", "ocl"] +allmodules = ["core", "flann", "imgproc", "ml", "highgui", "video", "features2d", "calib3d", "objdetect", "legacy", "contrib", "gpu", "androidcamera", "java", "python", "stitching", "ts", "photo", "nonfree", "videostab", "ocl", "superres"] verbose = False show_warnings = True show_errors = True @@ -380,7 +380,7 @@ class RstParser(object): @classmethod def parse_namespace(cls, func, section_name): - known_namespaces = ["cv", "gpu", "flann"] + known_namespaces = ["cv", "gpu", "flann", "superres"] l = section_name.strip() for namespace in known_namespaces: if l.startswith(namespace + "::"): diff --git a/modules/superres/CMakeLists.txt b/modules/superres/CMakeLists.txt new file mode 100644 index 0000000..5e82629 --- /dev/null +++ b/modules/superres/CMakeLists.txt @@ -0,0 +1,33 @@ +if(ANDROID OR IOS) + ocv_module_disable(superres) +endif() + +set(the_description "Super Resolution") +ocv_add_module(superres opencv_imgproc opencv_video OPTIONAL opencv_gpu opencv_highgui) +ocv_module_include_directories() + +ocv_warnings_disable(CMAKE_CXX_FLAGS -Wundef /wd4127) + +if(HAVE_CUDA) + string(REPLACE "-Wsign-promo" "" CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS}") + + ocv_source_group("Src\\Cuda" GLOB "src/cuda/*.cu") + ocv_include_directories("${OpenCV_SOURCE_DIR}/modules/gpu/include" ${CUDA_INCLUDE_DIRS}) + + file(GLOB lib_cuda "src/cuda/*.cu") + ocv_cuda_compile(cuda_objs ${lib_cuda}) + + set(cuda_link_libs ${CUDA_LIBRARIES}) +else() + set(lib_cuda "") + set(cuda_objs "") + set(cuda_link_libs "") +endif() + +ocv_glob_module_sources(SOURCES ${lib_cuda} ${cuda_objs}) + +ocv_create_module(${cuda_link_libs}) +ocv_add_precompiled_headers(${the_module}) + +ocv_add_accuracy_tests() +ocv_add_perf_tests() diff --git a/modules/superres/doc/super_resolution.rst b/modules/superres/doc/super_resolution.rst new file mode 100644 index 0000000..1772b0d --- /dev/null +++ b/modules/superres/doc/super_resolution.rst @@ -0,0 +1,84 @@ +Super Resolution +================ + +.. highlight:: cpp + +The Super Resolution module contains a set of functions and classes that can be used to solve the problem of resolution enhancement. There are a few methods implemented, most of them are descibed in the papers [Farsiu03]_ and [Mitzel09]_. + + + +superres::SuperResolution +------------------------- +Base class for Super Resolution algorithms. + +.. ocv:class:: superres::SuperResolution : public Algorithm, public superres::FrameSource + +The class is only used to define the common interface for the whole family of Super Resolution algorithms. + + + +superres::SuperResolution::setInput +----------------------------------- +Set input frame source for Super Resolution algorithm. + +.. ocv:function:: void superres::SuperResolution::setInput(const Ptr& frameSource) + + :param frameSource: Input frame source + + + +superres::SuperResolution::nextFrame +------------------------------------ +Process next frame from input and return output result. + +.. ocv:function:: void superres::SuperResolution::nextFrame(OutputArray frame) + + :param frame: Output result + + + +superres::SuperResolution::collectGarbage +----------------------------------------- +Clear all inner buffers. + +.. ocv:function:: void superres::SuperResolution::collectGarbage() + + + +superres::createSuperResolution_BTVL1 +------------------------------------- +Create Bilateral TV-L1 Super Resolution. + +.. ocv:function:: Ptr superres::createSuperResolution_BTVL1() + +.. ocv:function:: Ptr superres::createSuperResolution_BTVL1_GPU() + +This class implements Super Resolution algorithm described in the papers [Farsiu03]_ and [Mitzel09]_ . + +Here are important members of the class that control the algorithm, which you can set after constructing the class instance: + + * **int scale** Scale factor. + + * **int iterations** Iteration count. + + * **double tau** Asymptotic value of steepest descent method. + + * **double lambda** Weight parameter to balance data term and smoothness term. + + * **double alpha** Parameter of spacial distribution in Bilateral-TV. + + * **int btvKernelSize** Kernel size of Bilateral-TV filter. + + * **int blurKernelSize** Gaussian blur kernel size. + + * **double blurSigma** Gaussian blur sigma. + + * **int temporalAreaRadius** Radius of the temporal search area. + + * **Ptr opticalFlow** Dense optical flow algorithm. + + + +.. [Farsiu03] S. Farsiu, D. Robinson, M. Elad, P. Milanfar. Fast and robust Super-Resolution. Proc 2003 IEEE Int Conf on Image Process, pp. 291–294, 2003. + +.. [Mitzel09] D. Mitzel, T. Pock, T. Schoenemann, D. Cremers. Video super resolution using duality based TV-L1 optical flow. DAGM, 2009. diff --git a/modules/superres/doc/superres.rst b/modules/superres/doc/superres.rst new file mode 100644 index 0000000..6a69fca --- /dev/null +++ b/modules/superres/doc/superres.rst @@ -0,0 +1,8 @@ +************************** +superres. Super Resolution +************************** + +.. toctree:: + :maxdepth: 2 + + super_resolution diff --git a/modules/superres/include/opencv2/superres/optical_flow.hpp b/modules/superres/include/opencv2/superres/optical_flow.hpp new file mode 100644 index 0000000..bb344fc --- /dev/null +++ b/modules/superres/include/opencv2/superres/optical_flow.hpp @@ -0,0 +1,73 @@ +/*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) 2009-2011, 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*/ + +#ifndef __OPENCV_SUPERRES_OPTICAL_FLOW_HPP__ +#define __OPENCV_SUPERRES_OPTICAL_FLOW_HPP__ + +#include "opencv2/core/core.hpp" + +namespace cv +{ + namespace superres + { + class CV_EXPORTS DenseOpticalFlowExt : public cv::Algorithm + { + public: + virtual void calc(InputArray frame0, InputArray frame1, OutputArray flow1, OutputArray flow2 = noArray()) = 0; + virtual void collectGarbage() = 0; + }; + + CV_EXPORTS Ptr createOptFlow_Farneback(); + CV_EXPORTS Ptr createOptFlow_Farneback_GPU(); + + CV_EXPORTS Ptr createOptFlow_Simple(); + + CV_EXPORTS Ptr createOptFlow_DualTVL1(); + CV_EXPORTS Ptr createOptFlow_DualTVL1_GPU(); + + CV_EXPORTS Ptr createOptFlow_Brox_GPU(); + + CV_EXPORTS Ptr createOptFlow_PyrLK_GPU(); + } +} + +#endif // __OPENCV_SUPERRES_OPTICAL_FLOW_HPP__ diff --git a/modules/superres/include/opencv2/superres/superres.hpp b/modules/superres/include/opencv2/superres/superres.hpp new file mode 100644 index 0000000..e3e7a1e --- /dev/null +++ b/modules/superres/include/opencv2/superres/superres.hpp @@ -0,0 +1,98 @@ +/*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) 2009-2011, 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*/ + +#ifndef __OPENCV_SUPERRES_HPP__ +#define __OPENCV_SUPERRES_HPP__ + +#include "opencv2/core/core.hpp" + +namespace cv +{ + namespace superres + { + CV_EXPORTS bool initModule_superres(); + + class CV_EXPORTS FrameSource + { + public: + virtual ~FrameSource(); + + virtual void nextFrame(OutputArray frame) = 0; + virtual void reset() = 0; + }; + + CV_EXPORTS Ptr createFrameSource_Empty(); + + CV_EXPORTS Ptr createFrameSource_Video(const std::string& fileName); + CV_EXPORTS Ptr createFrameSource_Video_GPU(const std::string& fileName); + + CV_EXPORTS Ptr createFrameSource_Camera(int deviceId = 0); + + class CV_EXPORTS SuperResolution : public cv::Algorithm, public FrameSource + { + public: + void setInput(const Ptr& frameSource); + + void nextFrame(OutputArray frame); + void reset(); + + virtual void collectGarbage(); + + protected: + SuperResolution(); + + virtual void initImpl(Ptr& frameSource) = 0; + virtual void processImpl(Ptr& frameSource, OutputArray output) = 0; + + private: + Ptr frameSource_; + bool firstCall_; + }; + + // S. Farsiu , D. Robinson, M. Elad, P. Milanfar. Fast and robust multiframe super resolution. + // Dennis Mitzel, Thomas Pock, Thomas Schoenemann, Daniel Cremers. Video Super Resolution using Duality Based TV-L1 Optical Flow. + CV_EXPORTS Ptr createSuperResolution_BTVL1(); + CV_EXPORTS Ptr createSuperResolution_BTVL1_GPU(); + } +} + +#endif // __OPENCV_SUPERRES_HPP__ diff --git a/modules/superres/perf/perf_main.cpp b/modules/superres/perf/perf_main.cpp new file mode 100644 index 0000000..230df81 --- /dev/null +++ b/modules/superres/perf/perf_main.cpp @@ -0,0 +1,3 @@ +#include "perf_precomp.hpp" + +CV_PERF_TEST_MAIN(superres) diff --git a/modules/superres/perf/perf_precomp.cpp b/modules/superres/perf/perf_precomp.cpp new file mode 100644 index 0000000..8552ac3 --- /dev/null +++ b/modules/superres/perf/perf_precomp.cpp @@ -0,0 +1 @@ +#include "perf_precomp.hpp" diff --git a/modules/superres/perf/perf_precomp.hpp b/modules/superres/perf/perf_precomp.hpp new file mode 100644 index 0000000..732dd11 --- /dev/null +++ b/modules/superres/perf/perf_precomp.hpp @@ -0,0 +1,26 @@ +#ifdef __GNUC__ +# pragma GCC diagnostic ignored "-Wmissing-declarations" +# if defined __clang__ || defined __APPLE__ +# pragma GCC diagnostic ignored "-Wmissing-prototypes" +# pragma GCC diagnostic ignored "-Wextra" +# endif +#endif + +#ifndef __OPENCV_PERF_PRECOMP_HPP__ +#define __OPENCV_PERF_PRECOMP_HPP__ + +#ifdef HAVE_CVCONFIG_H +#include "cvconfig.h" +#endif + +#include "opencv2/core/core.hpp" +#include "opencv2/core/gpumat.hpp" +#include "opencv2/ts/ts_perf.hpp" +#include "opencv2/superres/superres.hpp" +#include "opencv2/superres/optical_flow.hpp" + +#ifdef GTEST_CREATE_SHARED_LIBRARY +#error no modules except ts should have GTEST_CREATE_SHARED_LIBRARY defined +#endif + +#endif diff --git a/modules/superres/perf/perf_superres.cpp b/modules/superres/perf/perf_superres.cpp new file mode 100644 index 0000000..a0fa8e5 --- /dev/null +++ b/modules/superres/perf/perf_superres.cpp @@ -0,0 +1,153 @@ +#include "perf_precomp.hpp" + +using namespace std; +using namespace std::tr1; +using namespace testing; +using namespace perf; +using namespace cv; +using namespace cv::superres; +using namespace cv::gpu; + +#define GPU_SANITY_CHECK(mat, ...) \ + do{ \ + Mat gpu_##mat(mat); \ + SANITY_CHECK(gpu_##mat, ## __VA_ARGS__); \ + } while(0) + +#define CPU_SANITY_CHECK(mat, ...) \ + do{ \ + Mat cpu_##mat(mat); \ + SANITY_CHECK(cpu_##mat, ## __VA_ARGS__); \ + } while(0) + +namespace +{ + class OneFrameSource_CPU : public FrameSource + { + public: + explicit OneFrameSource_CPU(const Mat& frame) : frame_(frame) {} + + void nextFrame(OutputArray frame) + { + frame.getMatRef() = frame_; + } + + void reset() + { + } + + private: + Mat frame_; + }; + + class OneFrameSource_GPU : public FrameSource + { + public: + explicit OneFrameSource_GPU(const GpuMat& frame) : frame_(frame) {} + + void nextFrame(OutputArray frame) + { + frame.getGpuMatRef() = frame_; + } + + void reset() + { + } + + private: + GpuMat frame_; + }; + + class ZeroOpticalFlow : public DenseOpticalFlowExt + { + public: + void calc(InputArray frame0, InputArray, OutputArray flow1, OutputArray flow2) + { + cv::Size size = frame0.size(); + + if (!flow2.needed()) + { + flow1.create(size, CV_32FC2); + + if (flow1.kind() == cv::_InputArray::GPU_MAT) + flow1.getGpuMatRef().setTo(cv::Scalar::all(0)); + else + flow1.getMatRef().setTo(cv::Scalar::all(0)); + } + else + { + flow1.create(size, CV_32FC1); + flow2.create(size, CV_32FC1); + + if (flow1.kind() == cv::_InputArray::GPU_MAT) + flow1.getGpuMatRef().setTo(cv::Scalar::all(0)); + else + flow1.getMatRef().setTo(cv::Scalar::all(0)); + + if (flow2.kind() == cv::_InputArray::GPU_MAT) + flow2.getGpuMatRef().setTo(cv::Scalar::all(0)); + else + flow2.getMatRef().setTo(cv::Scalar::all(0)); + } + } + + void collectGarbage() + { + } + }; +} + +PERF_TEST_P(Size_MatType, SuperResolution_BTVL1, + Combine(Values(szSmall64, szSmall128), + Values(MatType(CV_8UC1), MatType(CV_8UC3)))) +{ + declare.time(5 * 60); + + const Size size = get<0>(GetParam()); + const int type = get<1>(GetParam()); + + Mat frame(size, type); + declare.in(frame, WARMUP_RNG); + + const int scale = 2; + const int iterations = 50; + const int temporalAreaRadius = 1; + Ptr opticalFlow(new ZeroOpticalFlow); + + if (PERF_RUN_GPU()) + { + Ptr superRes = createSuperResolution_BTVL1_GPU(); + + superRes->set("scale", scale); + superRes->set("iterations", iterations); + superRes->set("temporalAreaRadius", temporalAreaRadius); + superRes->set("opticalFlow", opticalFlow); + + superRes->setInput(new OneFrameSource_GPU(GpuMat(frame))); + + GpuMat dst; + superRes->nextFrame(dst); + + TEST_CYCLE_N(10) superRes->nextFrame(dst); + + GPU_SANITY_CHECK(dst); + } + else + { + Ptr superRes = createSuperResolution_BTVL1(); + + superRes->set("scale", scale); + superRes->set("iterations", iterations); + superRes->set("temporalAreaRadius", temporalAreaRadius); + superRes->set("opticalFlow", opticalFlow); + + superRes->setInput(new OneFrameSource_CPU(frame)); + + Mat dst; + superRes->nextFrame(dst); + + TEST_CYCLE_N(10) superRes->nextFrame(dst); + + CPU_SANITY_CHECK(dst); + } +} diff --git a/modules/superres/src/btv_l1.cpp b/modules/superres/src/btv_l1.cpp new file mode 100644 index 0000000..71c6836 --- /dev/null +++ b/modules/superres/src/btv_l1.cpp @@ -0,0 +1,619 @@ +/*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) 2009-2011, 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*/ + +// S. Farsiu , D. Robinson, M. Elad, P. Milanfar. Fast and robust multiframe super resolution. +// Dennis Mitzel, Thomas Pock, Thomas Schoenemann, Daniel Cremers. Video Super Resolution using Duality Based TV-L1 Optical Flow. + +#include "precomp.hpp" + +using namespace std; +using namespace cv; +using namespace cv::superres; +using namespace cv::superres::detail; + +namespace +{ + void calcRelativeMotions(const vector& forwardMotions, const vector& backwardMotions, + vector& relForwardMotions, vector& relBackwardMotions, + int baseIdx, Size size) + { + const int count = static_cast(forwardMotions.size()); + + relForwardMotions.resize(count); + relForwardMotions[baseIdx].create(size, CV_32FC2); + relForwardMotions[baseIdx].setTo(Scalar::all(0)); + + relBackwardMotions.resize(count); + relBackwardMotions[baseIdx].create(size, CV_32FC2); + relBackwardMotions[baseIdx].setTo(Scalar::all(0)); + + for (int i = baseIdx - 1; i >= 0; --i) + { + add(relForwardMotions[i + 1], forwardMotions[i], relForwardMotions[i]); + + add(relBackwardMotions[i + 1], backwardMotions[i + 1], relBackwardMotions[i]); + } + + for (int i = baseIdx + 1; i < count; ++i) + { + add(relForwardMotions[i - 1], backwardMotions[i], relForwardMotions[i]); + + add(relBackwardMotions[i - 1], forwardMotions[i - 1], relBackwardMotions[i]); + } + } + + void upscaleMotions(const vector& lowResMotions, vector& highResMotions, int scale) + { + highResMotions.resize(lowResMotions.size()); + + for (size_t i = 0; i < lowResMotions.size(); ++i) + { + resize(lowResMotions[i], highResMotions[i], Size(), scale, scale, INTER_CUBIC); + multiply(highResMotions[i], Scalar::all(scale), highResMotions[i]); + } + } + + void buildMotionMaps(const Mat& forwardMotion, const Mat& backwardMotion, Mat& forwardMap, Mat& backwardMap) + { + forwardMap.create(forwardMotion.size(), CV_32FC2); + backwardMap.create(forwardMotion.size(), CV_32FC2); + + for (int y = 0; y < forwardMotion.rows; ++y) + { + const Point2f* forwardMotionRow = forwardMotion.ptr(y); + const Point2f* backwardMotionRow = backwardMotion.ptr(y); + Point2f* forwardMapRow = forwardMap.ptr(y); + Point2f* backwardMapRow = backwardMap.ptr(y); + + for (int x = 0; x < forwardMotion.cols; ++x) + { + Point2f base(static_cast(x), static_cast(y)); + + forwardMapRow[x] = base + backwardMotionRow[x]; + backwardMapRow[x] = base + forwardMotionRow[x]; + } + } + } + + template + void upscaleImpl(const Mat& src, Mat& dst, int scale) + { + dst.create(src.rows * scale, src.cols * scale, src.type()); + dst.setTo(Scalar::all(0)); + + for (int y = 0, Y = 0; y < src.rows; ++y, Y += scale) + { + const T* srcRow = src.ptr(y); + T* dstRow = dst.ptr(Y); + + for (int x = 0, X = 0; x < src.cols; ++x, X += scale) + dstRow[X] = srcRow[x]; + } + } + + void upscale(const Mat& src, Mat& dst, int scale) + { + typedef void (*func_t)(const Mat& src, Mat& dst, int scale); + static const func_t funcs[] = + { + 0, upscaleImpl, 0, upscaleImpl + }; + + CV_Assert( src.channels() == 1 || src.channels() == 3 || src.channels() == 4 ); + + const func_t func = funcs[src.channels()]; + + func(src, dst, scale); + } + + float diffSign(float a, float b) + { + return a > b ? 1.0f : a < b ? -1.0f : 0.0f; + } + Point3f diffSign(Point3f a, Point3f b) + { + return Point3f( + a.x > b.x ? 1.0f : a.x < b.x ? -1.0f : 0.0f, + a.y > b.y ? 1.0f : a.y < b.y ? -1.0f : 0.0f, + a.z > b.z ? 1.0f : a.z < b.z ? -1.0f : 0.0f + ); + } + + void diffSign(const Mat& src1, const Mat& src2, Mat& dst) + { + const int count = src1.cols * src1.channels(); + + dst.create(src1.size(), src1.type()); + + for (int y = 0; y < src1.rows; ++y) + { + const float* src1Ptr = src1.ptr(y); + const float* src2Ptr = src2.ptr(y); + float* dstPtr = dst.ptr(y); + + for (int x = 0; x < count; ++x) + dstPtr[x] = diffSign(src1Ptr[x], src2Ptr[x]); + } + } + + void calcBtvWeights(int btvKernelSize, double alpha, vector& btvWeights) + { + const size_t size = btvKernelSize * btvKernelSize; + + btvWeights.resize(size); + + const int ksize = (btvKernelSize - 1) / 2; + const float alpha_f = static_cast(alpha); + + for (int m = 0, ind = 0; m <= ksize; ++m) + { + for (int l = ksize; l + m >= 0; --l, ++ind) + btvWeights[ind] = pow(alpha_f, std::abs(m) + std::abs(l)); + } + } + + template + struct BtvRegularizationBody : ParallelLoopBody + { + void operator ()(const Range& range) const; + + Mat src; + mutable Mat dst; + int ksize; + const float* btvWeights; + }; + + template + void BtvRegularizationBody::operator ()(const Range& range) const + { + for (int i = range.start; i < range.end; ++i) + { + const T* srcRow = src.ptr(i); + T* dstRow = dst.ptr(i); + + for(int j = ksize; j < src.cols - ksize; ++j) + { + const T srcVal = srcRow[j]; + + for (int m = 0, ind = 0; m <= ksize; ++m) + { + const T* srcRow2 = src.ptr(i - m); + const T* srcRow3 = src.ptr(i + m); + + for (int l = ksize; l + m >= 0; --l, ++ind) + { + dstRow[j] += btvWeights[ind] * (diffSign(srcVal, srcRow3[j + l]) - diffSign(srcRow2[j - l], srcVal)); + } + } + } + } + } + + template + void calcBtvRegularizationImpl(const Mat& src, Mat& dst, int btvKernelSize, const vector& btvWeights) + { + dst.create(src.size(), src.type()); + dst.setTo(Scalar::all(0)); + + const int ksize = (btvKernelSize - 1) / 2; + + BtvRegularizationBody body; + + body.src = src; + body.dst = dst; + body.ksize = ksize; + body.btvWeights = &btvWeights[0]; + + parallel_for_(Range(ksize, src.rows - ksize), body); + } + + void calcBtvRegularization(const Mat& src, Mat& dst, int btvKernelSize, const vector& btvWeights) + { + typedef void (*func_t)(const Mat& src, Mat& dst, int btvKernelSize, const vector& btvWeights); + static const func_t funcs[] = + { + 0, calcBtvRegularizationImpl, 0, calcBtvRegularizationImpl + }; + + const func_t func = funcs[src.channels()]; + + func(src, dst, btvKernelSize, btvWeights); + } + + class BTVL1_Base + { + public: + BTVL1_Base(); + + void process(const vector& src, Mat& dst, + const vector& forwardMotions, const vector& backwardMotions, + int baseIdx); + + void collectGarbage(); + + protected: + int scale_; + int iterations_; + double tau_; + double lambda_; + double alpha_; + int btvKernelSize_; + int blurKernelSize_; + double blurSigma_; + Ptr opticalFlow_; + + private: + Ptr filter_; + int curBlurKernelSize_; + double curBlurSigma_; + int curSrcType_; + + vector btvWeights_; + int curBtvKernelSize_; + double curAlpha_; + + vector lowResForwardMotions_; + vector lowResBackwardMotions_; + + vector highResForwardMotions_; + vector highResBackwardMotions_; + + vector forwardMaps_; + vector backwardMaps_; + + Mat highRes_; + + Mat diffTerm_, regTerm_; + Mat a_, b_, c_; + }; + + BTVL1_Base::BTVL1_Base() + { + scale_ = 4; + iterations_ = 180; + lambda_ = 0.03; + tau_ = 1.3; + alpha_ = 0.7; + btvKernelSize_ = 7; + blurKernelSize_ = 5; + blurSigma_ = 0.0; + opticalFlow_ = createOptFlow_Farneback(); + + curBlurKernelSize_ = -1; + curBlurSigma_ = -1.0; + curSrcType_ = -1; + + curBtvKernelSize_ = -1; + curAlpha_ = -1.0; + } + + void BTVL1_Base::process(const vector& src, Mat& dst, const vector& forwardMotions, const vector& backwardMotions, int baseIdx) + { + CV_Assert( scale_ > 1 ); + CV_Assert( iterations_ > 0 ); + CV_Assert( tau_ > 0.0 ); + CV_Assert( alpha_ > 0.0 ); + CV_Assert( btvKernelSize_ > 0 ); + CV_Assert( blurKernelSize_ > 0 ); + CV_Assert( blurSigma_ >= 0.0 ); + + // update blur filter and btv weights + + if (filter_.empty() || blurKernelSize_ != curBlurKernelSize_ || blurSigma_ != curBlurSigma_ || src[0].type() != curSrcType_) + { + filter_ = createGaussianFilter(src[0].type(), Size(blurKernelSize_, blurKernelSize_), blurSigma_); + curBlurKernelSize_ = blurKernelSize_; + curBlurSigma_ = blurSigma_; + curSrcType_ = src[0].type(); + } + + if (btvWeights_.empty() || btvKernelSize_ != curBtvKernelSize_ || alpha_ != curAlpha_) + { + calcBtvWeights(btvKernelSize_, alpha_, btvWeights_); + curBtvKernelSize_ = btvKernelSize_; + curAlpha_ = alpha_; + } + + // calc high res motions + + calcRelativeMotions(forwardMotions, backwardMotions, lowResForwardMotions_, lowResBackwardMotions_, baseIdx, src[0].size()); + + upscaleMotions(lowResForwardMotions_, highResForwardMotions_, scale_); + upscaleMotions(lowResBackwardMotions_, highResBackwardMotions_, scale_); + + forwardMaps_.resize(highResForwardMotions_.size()); + backwardMaps_.resize(highResForwardMotions_.size()); + for (size_t i = 0; i < highResForwardMotions_.size(); ++i) + buildMotionMaps(highResForwardMotions_[i], highResBackwardMotions_[i], forwardMaps_[i], backwardMaps_[i]); + + // initial estimation + + const Size lowResSize = src[0].size(); + const Size highResSize(lowResSize.width * scale_, lowResSize.height * scale_); + + resize(src[baseIdx], highRes_, highResSize, 0, 0, INTER_CUBIC); + + // iterations + + diffTerm_.create(highResSize, highRes_.type()); + a_.create(highResSize, highRes_.type()); + b_.create(highResSize, highRes_.type()); + c_.create(lowResSize, highRes_.type()); + + for (int i = 0; i < iterations_; ++i) + { + diffTerm_.setTo(Scalar::all(0)); + + for (size_t k = 0; k < src.size(); ++k) + { + // a = M * Ih + remap(highRes_, a_, backwardMaps_[k], noArray(), INTER_NEAREST); + // b = HM * Ih + filter_->apply(a_, b_); + // c = DHM * Ih + resize(b_, c_, lowResSize, 0, 0, INTER_NEAREST); + + diffSign(src[k], c_, c_); + + // a = Dt * diff + upscale(c_, a_, scale_); + // b = HtDt * diff + filter_->apply(a_, b_); + // a = MtHtDt * diff + remap(b_, a_, forwardMaps_[k], noArray(), INTER_NEAREST); + + add(diffTerm_, a_, diffTerm_); + } + + if (lambda_ > 0) + { + calcBtvRegularization(highRes_, regTerm_, btvKernelSize_, btvWeights_); + addWeighted(diffTerm_, 1.0, regTerm_, -lambda_, 0.0, diffTerm_); + } + + addWeighted(highRes_, 1.0, diffTerm_, tau_, 0.0, highRes_); + } + + Rect inner(btvKernelSize_, btvKernelSize_, highRes_.cols - 2 * btvKernelSize_, highRes_.rows - 2 * btvKernelSize_); + highRes_(inner).copyTo(dst); + } + + void BTVL1_Base::collectGarbage() + { + filter_.release(); + + lowResForwardMotions_.clear(); + lowResBackwardMotions_.clear(); + + highResForwardMotions_.clear(); + highResBackwardMotions_.clear(); + + forwardMaps_.clear(); + backwardMaps_.clear(); + + highRes_.release(); + + diffTerm_.release(); + regTerm_.release(); + a_.release(); + b_.release(); + c_.release(); + } + +//////////////////////////////////////////////////////////////////// + + class BTVL1 : public SuperResolution, private BTVL1_Base + { + public: + AlgorithmInfo* info() const; + + BTVL1(); + + void collectGarbage(); + + protected: + void initImpl(Ptr& frameSource); + void processImpl(Ptr& frameSource, OutputArray output); + + private: + int temporalAreaRadius_; + + void readNextFrame(Ptr& frameSource); + void processFrame(int idx); + + Mat curFrame_; + Mat prevFrame_; + + vector frames_; + vector forwardMotions_; + vector backwardMotions_; + vector outputs_; + + int storePos_; + int procPos_; + int outPos_; + + vector srcFrames_; + vector srcForwardMotions_; + vector srcBackwardMotions_; + Mat finalOutput_; + }; + + CV_INIT_ALGORITHM(BTVL1, "SuperResolution.BTVL1", + obj.info()->addParam(obj, "scale", obj.scale_, false, 0, 0, "Scale factor."); + obj.info()->addParam(obj, "iterations", obj.iterations_, false, 0, 0, "Iteration count."); + obj.info()->addParam(obj, "tau", obj.tau_, false, 0, 0, "Asymptotic value of steepest descent method."); + obj.info()->addParam(obj, "lambda", obj.lambda_, false, 0, 0, "Weight parameter to balance data term and smoothness term."); + obj.info()->addParam(obj, "alpha", obj.alpha_, false, 0, 0, "Parameter of spacial distribution in Bilateral-TV."); + obj.info()->addParam(obj, "btvKernelSize", obj.btvKernelSize_, false, 0, 0, "Kernel size of Bilateral-TV filter."); + obj.info()->addParam(obj, "blurKernelSize", obj.blurKernelSize_, false, 0, 0, "Gaussian blur kernel size."); + obj.info()->addParam(obj, "blurSigma", obj.blurSigma_, false, 0, 0, "Gaussian blur sigma."); + obj.info()->addParam(obj, "temporalAreaRadius", obj.temporalAreaRadius_, false, 0, 0, "Radius of the temporal search area."); + obj.info()->addParam(obj, "opticalFlow", obj.opticalFlow_, false, 0, 0, "Dense optical flow algorithm.")); + + BTVL1::BTVL1() + { + temporalAreaRadius_ = 4; + } + + void BTVL1::collectGarbage() + { + curFrame_.release(); + prevFrame_.release(); + + frames_.clear(); + forwardMotions_.clear(); + backwardMotions_.clear(); + outputs_.clear(); + + srcFrames_.clear(); + srcForwardMotions_.clear(); + srcBackwardMotions_.clear(); + finalOutput_.release(); + + SuperResolution::collectGarbage(); + BTVL1_Base::collectGarbage(); + } + + void BTVL1::initImpl(Ptr& frameSource) + { + const int cacheSize = 2 * temporalAreaRadius_ + 1; + + frames_.resize(cacheSize); + forwardMotions_.resize(cacheSize); + backwardMotions_.resize(cacheSize); + outputs_.resize(cacheSize); + + storePos_ = -1; + + for (int t = -temporalAreaRadius_; t <= temporalAreaRadius_; ++t) + readNextFrame(frameSource); + + for (int i = 0; i <= temporalAreaRadius_; ++i) + processFrame(i); + + procPos_ = temporalAreaRadius_; + outPos_ = -1; + } + + void BTVL1::processImpl(Ptr& frameSource, OutputArray _output) + { + if (outPos_ >= storePos_) + { + _output.release(); + return; + } + + readNextFrame(frameSource); + + if (procPos_ < storePos_) + { + ++procPos_; + processFrame(procPos_); + } + + ++outPos_; + const Mat& curOutput = at(outPos_, outputs_); + + if (_output.kind() < _InputArray::OPENGL_BUFFER) + curOutput.convertTo(_output, CV_8U); + else + { + curOutput.convertTo(finalOutput_, CV_8U); + arrCopy(finalOutput_, _output); + } + } + + void BTVL1::readNextFrame(Ptr& frameSource) + { + frameSource->nextFrame(curFrame_); + + if (curFrame_.empty()) + return; + + ++storePos_; + curFrame_.convertTo(at(storePos_, frames_), CV_32F); + + if (storePos_ > 0) + { + opticalFlow_->calc(prevFrame_, curFrame_, at(storePos_ - 1, forwardMotions_)); + opticalFlow_->calc(curFrame_, prevFrame_, at(storePos_, backwardMotions_)); + } + + curFrame_.copyTo(prevFrame_); + } + + void BTVL1::processFrame(int idx) + { + const int startIdx = max(idx - temporalAreaRadius_, 0); + const int procIdx = idx; + const int endIdx = min(startIdx + 2 * temporalAreaRadius_, storePos_); + + const int count = endIdx - startIdx + 1; + + srcFrames_.resize(count); + srcForwardMotions_.resize(count); + srcBackwardMotions_.resize(count); + + int baseIdx = -1; + + for (int i = startIdx, k = 0; i <= endIdx; ++i, ++k) + { + if (i == procIdx) + baseIdx = k; + + srcFrames_[k] = at(i, frames_); + + if (i < endIdx) + srcForwardMotions_[k] = at(i, forwardMotions_); + if (i > startIdx) + srcBackwardMotions_[k] = at(i, backwardMotions_); + } + + process(srcFrames_, at(idx, outputs_), srcForwardMotions_, srcBackwardMotions_, baseIdx); + } +} + +Ptr cv::superres::createSuperResolution_BTVL1() +{ + return new BTVL1; +} diff --git a/modules/superres/src/btv_l1_gpu.cpp b/modules/superres/src/btv_l1_gpu.cpp new file mode 100644 index 0000000..4820d71 --- /dev/null +++ b/modules/superres/src/btv_l1_gpu.cpp @@ -0,0 +1,580 @@ +/*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) 2009-2011, 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*/ + +// S. Farsiu , D. Robinson, M. Elad, P. Milanfar. Fast and robust multiframe super resolution. +// Dennis Mitzel, Thomas Pock, Thomas Schoenemann, Daniel Cremers. Video Super Resolution using Duality Based TV-L1 Optical Flow. + +#include "precomp.hpp" + +using namespace std; +using namespace cv; +using namespace cv::gpu; +using namespace cv::superres; +using namespace cv::superres::detail; + +#ifndef HAVE_CUDA + +Ptr cv::superres::createSuperResolution_BTVL1_GPU() +{ + CV_Error(CV_StsNotImplemented, "The called functionality is disabled for current build or platform"); + return Ptr(); +} + +#else // HAVE_CUDA + +namespace btv_l1_device +{ + void buildMotionMaps(PtrStepSzf forwardMotionX, PtrStepSzf forwardMotionY, + PtrStepSzf backwardMotionX, PtrStepSzf bacwardMotionY, + PtrStepSzf forwardMapX, PtrStepSzf forwardMapY, + PtrStepSzf backwardMapX, PtrStepSzf backwardMapY); + + template + void upscale(const PtrStepSzb src, PtrStepSzb dst, int scale, cudaStream_t stream); + + void diffSign(PtrStepSzf src1, PtrStepSzf src2, PtrStepSzf dst, cudaStream_t stream); + + void loadBtvWeights(const float* weights, size_t count); + template void calcBtvRegularization(PtrStepSzb src, PtrStepSzb dst, int ksize); +} + +namespace +{ + void calcRelativeMotions(const vector >& forwardMotions, const vector >& backwardMotions, + vector >& relForwardMotions, vector >& relBackwardMotions, + int baseIdx, Size size) + { + const int count = static_cast(forwardMotions.size()); + + relForwardMotions.resize(count); + relForwardMotions[baseIdx].first.create(size, CV_32FC1); + relForwardMotions[baseIdx].first.setTo(Scalar::all(0)); + relForwardMotions[baseIdx].second.create(size, CV_32FC1); + relForwardMotions[baseIdx].second.setTo(Scalar::all(0)); + + relBackwardMotions.resize(count); + relBackwardMotions[baseIdx].first.create(size, CV_32FC1); + relBackwardMotions[baseIdx].first.setTo(Scalar::all(0)); + relBackwardMotions[baseIdx].second.create(size, CV_32FC1); + relBackwardMotions[baseIdx].second.setTo(Scalar::all(0)); + + for (int i = baseIdx - 1; i >= 0; --i) + { + gpu::add(relForwardMotions[i + 1].first, forwardMotions[i].first, relForwardMotions[i].first); + gpu::add(relForwardMotions[i + 1].second, forwardMotions[i].second, relForwardMotions[i].second); + + gpu::add(relBackwardMotions[i + 1].first, backwardMotions[i + 1].first, relBackwardMotions[i].first); + gpu::add(relBackwardMotions[i + 1].second, backwardMotions[i + 1].second, relBackwardMotions[i].second); + } + + for (int i = baseIdx + 1; i < count; ++i) + { + gpu::add(relForwardMotions[i - 1].first, backwardMotions[i].first, relForwardMotions[i].first); + gpu::add(relForwardMotions[i - 1].second, backwardMotions[i].second, relForwardMotions[i].second); + + gpu::add(relBackwardMotions[i - 1].first, forwardMotions[i - 1].first, relBackwardMotions[i].first); + gpu::add(relBackwardMotions[i - 1].second, forwardMotions[i - 1].second, relBackwardMotions[i].second); + } + } + + void upscaleMotions(const vector >& lowResMotions, vector >& highResMotions, int scale) + { + highResMotions.resize(lowResMotions.size()); + + for (size_t i = 0; i < lowResMotions.size(); ++i) + { + gpu::resize(lowResMotions[i].first, highResMotions[i].first, Size(), scale, scale, INTER_CUBIC); + gpu::resize(lowResMotions[i].second, highResMotions[i].second, Size(), scale, scale, INTER_CUBIC); + + gpu::multiply(highResMotions[i].first, Scalar::all(scale), highResMotions[i].first); + gpu::multiply(highResMotions[i].second, Scalar::all(scale), highResMotions[i].second); + } + } + + void buildMotionMaps(const pair& forwardMotion, const pair& backwardMotion, + pair& forwardMap, pair& backwardMap) + { + forwardMap.first.create(forwardMotion.first.size(), CV_32FC1); + forwardMap.second.create(forwardMotion.first.size(), CV_32FC1); + + backwardMap.first.create(forwardMotion.first.size(), CV_32FC1); + backwardMap.second.create(forwardMotion.first.size(), CV_32FC1); + + btv_l1_device::buildMotionMaps(forwardMotion.first, forwardMotion.second, + backwardMotion.first, backwardMotion.second, + forwardMap.first, forwardMap.second, + backwardMap.first, backwardMap.second); + } + + void upscale(const GpuMat& src, GpuMat& dst, int scale, Stream& stream) + { + typedef void (*func_t)(const PtrStepSzb src, PtrStepSzb dst, int scale, cudaStream_t stream); + static const func_t funcs[] = + { + 0, btv_l1_device::upscale<1>, 0, btv_l1_device::upscale<3>, btv_l1_device::upscale<4> + }; + + CV_Assert( src.channels() == 1 || src.channels() == 3 || src.channels() == 4 ); + + dst.create(src.rows * scale, src.cols * scale, src.type()); + dst.setTo(Scalar::all(0)); + + const func_t func = funcs[src.channels()]; + + func(src, dst, scale, StreamAccessor::getStream(stream)); + } + + void diffSign(const GpuMat& src1, const GpuMat& src2, GpuMat& dst, Stream& stream) + { + dst.create(src1.size(), src1.type()); + + btv_l1_device::diffSign(src1.reshape(1), src2.reshape(1), dst.reshape(1), StreamAccessor::getStream(stream)); + } + + void calcBtvWeights(int btvKernelSize, double alpha, vector& btvWeights) + { + const size_t size = btvKernelSize * btvKernelSize; + + btvWeights.resize(size); + + const int ksize = (btvKernelSize - 1) / 2; + const float alpha_f = static_cast(alpha); + + for (int m = 0, ind = 0; m <= ksize; ++m) + { + for (int l = ksize; l + m >= 0; --l, ++ind) + btvWeights[ind] = pow(alpha_f, std::abs(m) + std::abs(l)); + } + + btv_l1_device::loadBtvWeights(&btvWeights[0], size); + } + + void calcBtvRegularization(const GpuMat& src, GpuMat& dst, int btvKernelSize) + { + typedef void (*func_t)(PtrStepSzb src, PtrStepSzb dst, int ksize); + static const func_t funcs[] = + { + 0, + btv_l1_device::calcBtvRegularization<1>, + 0, + btv_l1_device::calcBtvRegularization<3>, + btv_l1_device::calcBtvRegularization<4> + }; + + dst.create(src.size(), src.type()); + dst.setTo(Scalar::all(0)); + + const int ksize = (btvKernelSize - 1) / 2; + + funcs[src.channels()](src, dst, ksize); + } + + class BTVL1_GPU_Base + { + public: + BTVL1_GPU_Base(); + + void process(const vector& src, GpuMat& dst, + const vector >& forwardMotions, const vector >& backwardMotions, + int baseIdx); + + void collectGarbage(); + + protected: + int scale_; + int iterations_; + double lambda_; + double tau_; + double alpha_; + int btvKernelSize_; + int blurKernelSize_; + double blurSigma_; + Ptr opticalFlow_; + + private: + vector > filters_; + int curBlurKernelSize_; + double curBlurSigma_; + int curSrcType_; + + vector btvWeights_; + int curBtvKernelSize_; + double curAlpha_; + + vector > lowResForwardMotions_; + vector > lowResBackwardMotions_; + + vector > highResForwardMotions_; + vector > highResBackwardMotions_; + + vector > forwardMaps_; + vector > backwardMaps_; + + GpuMat highRes_; + + vector streams_; + vector diffTerms_; + vector a_, b_, c_; + GpuMat regTerm_; + }; + + BTVL1_GPU_Base::BTVL1_GPU_Base() + { + scale_ = 4; + iterations_ = 180; + lambda_ = 0.03; + tau_ = 1.3; + alpha_ = 0.7; + btvKernelSize_ = 7; + blurKernelSize_ = 5; + blurSigma_ = 0.0; + opticalFlow_ = createOptFlow_Farneback_GPU(); + + curBlurKernelSize_ = -1; + curBlurSigma_ = -1.0; + curSrcType_ = -1; + + curBtvKernelSize_ = -1; + curAlpha_ = -1.0; + } + + void BTVL1_GPU_Base::process(const vector& src, GpuMat& dst, + const vector >& forwardMotions, const vector >& backwardMotions, + int baseIdx) + { + CV_Assert( scale_ > 1 ); + CV_Assert( iterations_ > 0 ); + CV_Assert( tau_ > 0.0 ); + CV_Assert( alpha_ > 0.0 ); + CV_Assert( btvKernelSize_ > 0 && btvKernelSize_ <= 16 ); + CV_Assert( blurKernelSize_ > 0 ); + CV_Assert( blurSigma_ >= 0.0 ); + + // update blur filter and btv weights + + if (filters_.size() != src.size() || blurKernelSize_ != curBlurKernelSize_ || blurSigma_ != curBlurSigma_ || src[0].type() != curSrcType_) + { + filters_.resize(src.size()); + for (size_t i = 0; i < src.size(); ++i) + filters_[i] = createGaussianFilter_GPU(src[0].type(), Size(blurKernelSize_, blurKernelSize_), blurSigma_); + curBlurKernelSize_ = blurKernelSize_; + curBlurSigma_ = blurSigma_; + curSrcType_ = src[0].type(); + } + + if (btvWeights_.empty() || btvKernelSize_ != curBtvKernelSize_ || alpha_ != curAlpha_) + { + calcBtvWeights(btvKernelSize_, alpha_, btvWeights_); + curBtvKernelSize_ = btvKernelSize_; + curAlpha_ = alpha_; + } + + // calc motions between input frames + + calcRelativeMotions(forwardMotions, backwardMotions, lowResForwardMotions_, lowResBackwardMotions_, baseIdx, src[0].size()); + + upscaleMotions(lowResForwardMotions_, highResForwardMotions_, scale_); + upscaleMotions(lowResBackwardMotions_, highResBackwardMotions_, scale_); + + forwardMaps_.resize(highResForwardMotions_.size()); + backwardMaps_.resize(highResForwardMotions_.size()); + for (size_t i = 0; i < highResForwardMotions_.size(); ++i) + buildMotionMaps(highResForwardMotions_[i], highResBackwardMotions_[i], forwardMaps_[i], backwardMaps_[i]); + + // initial estimation + + const Size lowResSize = src[0].size(); + const Size highResSize(lowResSize.width * scale_, lowResSize.height * scale_); + + gpu::resize(src[baseIdx], highRes_, highResSize, 0, 0, INTER_CUBIC); + + // iterations + + streams_.resize(src.size()); + diffTerms_.resize(src.size()); + a_.resize(src.size()); + b_.resize(src.size()); + c_.resize(src.size()); + + for (int i = 0; i < iterations_; ++i) + { + for (size_t k = 0; k < src.size(); ++k) + { + // a = M * Ih + gpu::remap(highRes_, a_[k], backwardMaps_[k].first, backwardMaps_[k].second, INTER_NEAREST, BORDER_REPLICATE, Scalar(), streams_[k]); + // b = HM * Ih + filters_[k]->apply(a_[k], b_[k], Rect(0,0,-1,-1), streams_[k]); + // c = DHF * Ih + gpu::resize(b_[k], c_[k], lowResSize, 0, 0, INTER_NEAREST, streams_[k]); + + diffSign(src[k], c_[k], c_[k], streams_[k]); + + // a = Dt * diff + upscale(c_[k], a_[k], scale_, streams_[k]); + // b = HtDt * diff + filters_[k]->apply(a_[k], b_[k], Rect(0,0,-1,-1), streams_[k]); + // diffTerm = MtHtDt * diff + gpu::remap(b_[k], diffTerms_[k], forwardMaps_[k].first, forwardMaps_[k].second, INTER_NEAREST, BORDER_REPLICATE, Scalar(), streams_[k]); + } + + if (lambda_ > 0) + { + calcBtvRegularization(highRes_, regTerm_, btvKernelSize_); + gpu::addWeighted(highRes_, 1.0, regTerm_, -tau_ * lambda_, 0.0, highRes_); + } + + for (size_t k = 0; k < src.size(); ++k) + { + streams_[k].waitForCompletion(); + gpu::addWeighted(highRes_, 1.0, diffTerms_[k], tau_, 0.0, highRes_); + } + } + + Rect inner(btvKernelSize_, btvKernelSize_, highRes_.cols - 2 * btvKernelSize_, highRes_.rows - 2 * btvKernelSize_); + highRes_(inner).copyTo(dst); + } + + void BTVL1_GPU_Base::collectGarbage() + { + filters_.clear(); + + lowResForwardMotions_.clear(); + lowResBackwardMotions_.clear(); + + highResForwardMotions_.clear(); + highResBackwardMotions_.clear(); + + forwardMaps_.clear(); + backwardMaps_.clear(); + + highRes_.release(); + + diffTerms_.clear(); + a_.clear(); + b_.clear(); + c_.clear(); + regTerm_.release(); + } + +//////////////////////////////////////////////////////////// + + class BTVL1_GPU : public SuperResolution, private BTVL1_GPU_Base + { + public: + AlgorithmInfo* info() const; + + BTVL1_GPU(); + + void collectGarbage(); + + protected: + void initImpl(Ptr& frameSource); + void processImpl(Ptr& frameSource, OutputArray output); + + private: + int temporalAreaRadius_; + + void readNextFrame(Ptr& frameSource); + void processFrame(int idx); + + GpuMat curFrame_; + GpuMat prevFrame_; + + vector frames_; + vector > forwardMotions_; + vector > backwardMotions_; + vector outputs_; + + int storePos_; + int procPos_; + int outPos_; + + vector srcFrames_; + vector > srcForwardMotions_; + vector > srcBackwardMotions_; + GpuMat finalOutput_; + }; + + CV_INIT_ALGORITHM(BTVL1_GPU, "SuperResolution.BTVL1_GPU", + obj.info()->addParam(obj, "scale", obj.scale_, false, 0, 0, "Scale factor."); + obj.info()->addParam(obj, "iterations", obj.iterations_, false, 0, 0, "Iteration count."); + obj.info()->addParam(obj, "tau", obj.tau_, false, 0, 0, "Asymptotic value of steepest descent method."); + obj.info()->addParam(obj, "lambda", obj.lambda_, false, 0, 0, "Weight parameter to balance data term and smoothness term."); + obj.info()->addParam(obj, "alpha", obj.alpha_, false, 0, 0, "Parameter of spacial distribution in Bilateral-TV."); + obj.info()->addParam(obj, "btvKernelSize", obj.btvKernelSize_, false, 0, 0, "Kernel size of Bilateral-TV filter."); + obj.info()->addParam(obj, "blurKernelSize", obj.blurKernelSize_, false, 0, 0, "Gaussian blur kernel size."); + obj.info()->addParam(obj, "blurSigma", obj.blurSigma_, false, 0, 0, "Gaussian blur sigma."); + obj.info()->addParam(obj, "temporalAreaRadius", obj.temporalAreaRadius_, false, 0, 0, "Radius of the temporal search area."); + obj.info()->addParam(obj, "opticalFlow", obj.opticalFlow_, false, 0, 0, "Dense optical flow algorithm.")); + + BTVL1_GPU::BTVL1_GPU() + { + temporalAreaRadius_ = 4; + } + + void BTVL1_GPU::collectGarbage() + { + curFrame_.release(); + prevFrame_.release(); + + frames_.clear(); + forwardMotions_.clear(); + backwardMotions_.clear(); + outputs_.clear(); + + srcFrames_.clear(); + srcForwardMotions_.clear(); + srcBackwardMotions_.clear(); + finalOutput_.release(); + + SuperResolution::collectGarbage(); + BTVL1_GPU_Base::collectGarbage(); + } + + void BTVL1_GPU::initImpl(Ptr& frameSource) + { + const int cacheSize = 2 * temporalAreaRadius_ + 1; + + frames_.resize(cacheSize); + forwardMotions_.resize(cacheSize); + backwardMotions_.resize(cacheSize); + outputs_.resize(cacheSize); + + storePos_ = -1; + + for (int t = -temporalAreaRadius_; t <= temporalAreaRadius_; ++t) + readNextFrame(frameSource); + + for (int i = 0; i <= temporalAreaRadius_; ++i) + processFrame(i); + + procPos_ = temporalAreaRadius_; + outPos_ = -1; + } + + void BTVL1_GPU::processImpl(Ptr& frameSource, OutputArray _output) + { + if (outPos_ >= storePos_) + { + _output.release(); + return; + } + + readNextFrame(frameSource); + + if (procPos_ < storePos_) + { + ++procPos_; + processFrame(procPos_); + } + + ++outPos_; + const GpuMat& curOutput = at(outPos_, outputs_); + + if (_output.kind() == _InputArray::GPU_MAT) + curOutput.convertTo(_output.getGpuMatRef(), CV_8U); + else + { + curOutput.convertTo(finalOutput_, CV_8U); + arrCopy(finalOutput_, _output); + } + } + + void BTVL1_GPU::readNextFrame(Ptr& frameSource) + { + frameSource->nextFrame(curFrame_); + + if (curFrame_.empty()) + return; + + ++storePos_; + curFrame_.convertTo(at(storePos_, frames_), CV_32F); + + if (storePos_ > 0) + { + pair& forwardMotion = at(storePos_ - 1, forwardMotions_); + pair& backwardMotion = at(storePos_, backwardMotions_); + + opticalFlow_->calc(prevFrame_, curFrame_, forwardMotion.first, forwardMotion.second); + opticalFlow_->calc(curFrame_, prevFrame_, backwardMotion.first, backwardMotion.second); + } + + curFrame_.copyTo(prevFrame_); + } + + void BTVL1_GPU::processFrame(int idx) + { + const int startIdx = max(idx - temporalAreaRadius_, 0); + const int procIdx = idx; + const int endIdx = min(startIdx + 2 * temporalAreaRadius_, storePos_); + + const int count = endIdx - startIdx + 1; + + srcFrames_.resize(count); + srcForwardMotions_.resize(count); + srcBackwardMotions_.resize(count); + + int baseIdx = -1; + + for (int i = startIdx, k = 0; i <= endIdx; ++i, ++k) + { + if (i == procIdx) + baseIdx = k; + + srcFrames_[k] = at(i, frames_); + + if (i < endIdx) + srcForwardMotions_[k] = at(i, forwardMotions_); + if (i > startIdx) + srcBackwardMotions_[k] = at(i, backwardMotions_); + } + + process(srcFrames_, at(idx, outputs_), srcForwardMotions_, srcBackwardMotions_, baseIdx); + } +} + +Ptr cv::superres::createSuperResolution_BTVL1_GPU() +{ + return new BTVL1_GPU; +} + +#endif // HAVE_CUDA diff --git a/modules/superres/src/cuda/btv_l1_gpu.cu b/modules/superres/src/cuda/btv_l1_gpu.cu new file mode 100644 index 0000000..772e11d --- /dev/null +++ b/modules/superres/src/cuda/btv_l1_gpu.cu @@ -0,0 +1,234 @@ +/*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) 2009-2011, 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 "opencv2/gpu/device/common.hpp" +#include "opencv2/gpu/device/transform.hpp" +#include "opencv2/gpu/device/vec_traits.hpp" +#include "opencv2/gpu/device/vec_math.hpp" + +using namespace cv::gpu; +using namespace cv::gpu::device; + +namespace btv_l1_device +{ + void buildMotionMaps(PtrStepSzf forwardMotionX, PtrStepSzf forwardMotionY, + PtrStepSzf backwardMotionX, PtrStepSzf bacwardMotionY, + PtrStepSzf forwardMapX, PtrStepSzf forwardMapY, + PtrStepSzf backwardMapX, PtrStepSzf backwardMapY); + + template + void upscale(const PtrStepSzb src, PtrStepSzb dst, int scale, cudaStream_t stream); + + void diffSign(PtrStepSzf src1, PtrStepSzf src2, PtrStepSzf dst, cudaStream_t stream); + + void loadBtvWeights(const float* weights, size_t count); + template void calcBtvRegularization(PtrStepSzb src, PtrStepSzb dst, int ksize); +} + +namespace btv_l1_device +{ + __global__ void buildMotionMapsKernel(const PtrStepSzf forwardMotionX, const PtrStepf forwardMotionY, + PtrStepf backwardMotionX, PtrStepf backwardMotionY, + PtrStepf forwardMapX, PtrStepf forwardMapY, + PtrStepf backwardMapX, PtrStepf backwardMapY) + { + const int x = blockIdx.x * blockDim.x + threadIdx.x; + const int y = blockIdx.y * blockDim.y + threadIdx.y; + + if (x >= forwardMotionX.cols || y >= forwardMotionX.rows) + return; + + const float fx = forwardMotionX(y, x); + const float fy = forwardMotionY(y, x); + + const float bx = backwardMotionX(y, x); + const float by = backwardMotionY(y, x); + + forwardMapX(y, x) = x + bx; + forwardMapY(y, x) = y + by; + + backwardMapX(y, x) = x + fx; + backwardMapY(y, x) = y + fy; + } + + void buildMotionMaps(PtrStepSzf forwardMotionX, PtrStepSzf forwardMotionY, + PtrStepSzf backwardMotionX, PtrStepSzf bacwardMotionY, + PtrStepSzf forwardMapX, PtrStepSzf forwardMapY, + PtrStepSzf backwardMapX, PtrStepSzf backwardMapY) + { + const dim3 block(32, 8); + const dim3 grid(divUp(forwardMapX.cols, block.x), divUp(forwardMapX.rows, block.y)); + + buildMotionMapsKernel<<>>(forwardMotionX, forwardMotionY, + backwardMotionX, bacwardMotionY, + forwardMapX, forwardMapY, + backwardMapX, backwardMapY); + cudaSafeCall( cudaGetLastError() ); + + cudaSafeCall( cudaDeviceSynchronize() ); + } + + template + __global__ void upscaleKernel(const PtrStepSz src, PtrStep dst, const int scale) + { + const int x = blockIdx.x * blockDim.x + threadIdx.x; + const int y = blockIdx.y * blockDim.y + threadIdx.y; + + if (x >= src.cols || y >= src.rows) + return; + + dst(y * scale, x * scale) = src(y, x); + } + + template + void upscale(const PtrStepSzb src, PtrStepSzb dst, int scale, cudaStream_t stream) + { + typedef typename TypeVec::vec_type src_t; + + const dim3 block(32, 8); + const dim3 grid(divUp(src.cols, block.x), divUp(src.rows, block.y)); + + upscaleKernel<<>>((PtrStepSz) src, (PtrStepSz) dst, scale); + cudaSafeCall( cudaGetLastError() ); + + if (stream == 0) + cudaSafeCall( cudaDeviceSynchronize() ); + } + + template void upscale<1>(const PtrStepSzb src, PtrStepSzb dst, int scale, cudaStream_t stream); + template void upscale<3>(const PtrStepSzb src, PtrStepSzb dst, int scale, cudaStream_t stream); + template void upscale<4>(const PtrStepSzb src, PtrStepSzb dst, int scale, cudaStream_t stream); + + __device__ __forceinline__ float diffSign(float a, float b) + { + return a > b ? 1.0f : a < b ? -1.0f : 0.0f; + } + __device__ __forceinline__ float3 diffSign(const float3& a, const float3& b) + { + return make_float3( + a.x > b.x ? 1.0f : a.x < b.x ? -1.0f : 0.0f, + a.y > b.y ? 1.0f : a.y < b.y ? -1.0f : 0.0f, + a.z > b.z ? 1.0f : a.z < b.z ? -1.0f : 0.0f + ); + } + __device__ __forceinline__ float4 diffSign(const float4& a, const float4& b) + { + return make_float4( + a.x > b.x ? 1.0f : a.x < b.x ? -1.0f : 0.0f, + a.y > b.y ? 1.0f : a.y < b.y ? -1.0f : 0.0f, + a.z > b.z ? 1.0f : a.z < b.z ? -1.0f : 0.0f, + 0.0f + ); + } + + struct DiffSign : binary_function + { + __device__ __forceinline__ float operator ()(float a, float b) const + { + return diffSign(a, b); + } + }; +} + +namespace cv { namespace gpu { namespace device +{ + template <> struct TransformFunctorTraits : DefaultTransformFunctorTraits + { + enum { smart_block_dim_y = 8 }; + enum { smart_shift = 4 }; + }; +}}} + +namespace btv_l1_device +{ + void diffSign(PtrStepSzf src1, PtrStepSzf src2, PtrStepSzf dst, cudaStream_t stream) + { + transform(src1, src2, dst, DiffSign(), WithOutMask(), stream); + } + + __constant__ float c_btvRegWeights[16*16]; + + template + __global__ void calcBtvRegularizationKernel(const PtrStepSz src, PtrStep dst, const int ksize) + { + const int x = blockIdx.x * blockDim.x + threadIdx.x + ksize; + const int y = blockIdx.y * blockDim.y + threadIdx.y + ksize; + + if (y >= src.rows - ksize || x >= src.cols - ksize) + return; + + const T srcVal = src(y, x); + + T dstVal = VecTraits::all(0); + + for (int m = 0, count = 0; m <= ksize; ++m) + { + for (int l = ksize; l + m >= 0; --l, ++count) + dstVal = dstVal + c_btvRegWeights[count] * (diffSign(srcVal, src(y + m, x + l)) - diffSign(src(y - m, x - l), srcVal)); + } + + dst(y, x) = dstVal; + } + + void loadBtvWeights(const float* weights, size_t count) + { + cudaSafeCall( cudaMemcpyToSymbol(c_btvRegWeights, weights, count * sizeof(float)) ); + } + + template + void calcBtvRegularization(PtrStepSzb src, PtrStepSzb dst, int ksize) + { + typedef typename TypeVec::vec_type src_t; + + const dim3 block(32, 8); + const dim3 grid(divUp(src.cols, block.x), divUp(src.rows, block.y)); + + calcBtvRegularizationKernel<<>>((PtrStepSz) src, (PtrStepSz) dst, ksize); + cudaSafeCall( cudaGetLastError() ); + + cudaSafeCall( cudaDeviceSynchronize() ); + } + + template void calcBtvRegularization<1>(PtrStepSzb src, PtrStepSzb dst, int ksize); + template void calcBtvRegularization<3>(PtrStepSzb src, PtrStepSzb dst, int ksize); + template void calcBtvRegularization<4>(PtrStepSzb src, PtrStepSzb dst, int ksize); +} diff --git a/modules/superres/src/frame_source.cpp b/modules/superres/src/frame_source.cpp new file mode 100644 index 0000000..b22d0d0 --- /dev/null +++ b/modules/superres/src/frame_source.cpp @@ -0,0 +1,255 @@ +/*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) 2009-2011, 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" + +using namespace std; +using namespace cv; +using namespace cv::gpu; +using namespace cv::superres; +using namespace cv::superres::detail; + +cv::superres::FrameSource::~FrameSource() +{ +} + +////////////////////////////////////////////////////// +// EmptyFrameSource + +namespace +{ + class EmptyFrameSource : public FrameSource + { + public: + void nextFrame(OutputArray frame); + void reset(); + }; + + void EmptyFrameSource::nextFrame(OutputArray frame) + { + frame.release(); + } + + void EmptyFrameSource::reset() + { + } +} + +Ptr cv::superres::createFrameSource_Empty() +{ + return new EmptyFrameSource; +} + +////////////////////////////////////////////////////// +// VideoFrameSource & CameraFrameSource + +#ifndef HAVE_OPENCV_HIGHGUI + +Ptr cv::superres::createFrameSource_Video(const string& fileName) +{ + (void) fileName; + CV_Error(CV_StsNotImplemented, "The called functionality is disabled for current build or platform"); + return Ptr(); +} + +Ptr cv::superres::createFrameSource_Camera(int deviceId) +{ + (void) deviceId; + CV_Error(CV_StsNotImplemented, "The called functionality is disabled for current build or platform"); + return Ptr(); +} + +#else // HAVE_OPENCV_HIGHGUI + +namespace +{ + class CaptureFrameSource : public FrameSource + { + public: + void nextFrame(OutputArray frame); + + protected: + VideoCapture vc_; + + private: + Mat frame_; + }; + + void CaptureFrameSource::nextFrame(OutputArray _frame) + { + if (_frame.kind() == _InputArray::MAT) + { + vc_ >> _frame.getMatRef(); + } + else + { + vc_ >> frame_; + arrCopy(frame_, _frame); + } + } + + class VideoFrameSource : public CaptureFrameSource + { + public: + VideoFrameSource(const string& fileName); + + void reset(); + + private: + string fileName_; + }; + + VideoFrameSource::VideoFrameSource(const string& fileName) : fileName_(fileName) + { + reset(); + } + + void VideoFrameSource::reset() + { + vc_.release(); + vc_.open(fileName_); + CV_Assert( vc_.isOpened() ); + } + + class CameraFrameSource : public CaptureFrameSource + { + public: + CameraFrameSource(int deviceId); + + void reset(); + + private: + int deviceId_; + }; + + CameraFrameSource::CameraFrameSource(int deviceId) : deviceId_(deviceId) + { + reset(); + } + + void CameraFrameSource::reset() + { + vc_.release(); + vc_.open(deviceId_); + CV_Assert( vc_.isOpened() ); + } +} + +Ptr cv::superres::createFrameSource_Video(const string& fileName) +{ + return new VideoFrameSource(fileName); +} + +Ptr cv::superres::createFrameSource_Camera(int deviceId) +{ + return new CameraFrameSource(deviceId); +} + +#endif // HAVE_OPENCV_HIGHGUI + +////////////////////////////////////////////////////// +// VideoFrameSource_GPU + +#ifndef HAVE_OPENCV_GPU + +Ptr cv::superres::createFrameSource_Video_GPU(const string& fileName) +{ + (void) fileName; + CV_Error(CV_StsNotImplemented, "The called functionality is disabled for current build or platform"); + return Ptr(); +} + +#else // HAVE_OPENCV_GPU + +namespace +{ + class VideoFrameSource_GPU : public FrameSource + { + public: + VideoFrameSource_GPU(const string& fileName); + + void nextFrame(OutputArray frame); + void reset(); + + private: + string fileName_; + VideoReader_GPU reader_; + GpuMat frame_; + }; + + VideoFrameSource_GPU::VideoFrameSource_GPU(const string& fileName) : fileName_(fileName) + { + reset(); + } + + void VideoFrameSource_GPU::nextFrame(OutputArray _frame) + { + if (_frame.kind() == _InputArray::GPU_MAT) + { + bool res = reader_.read(_frame.getGpuMatRef()); + if (!res) + _frame.release(); + } + else + { + bool res = reader_.read(frame_); + if (!res) + _frame.release(); + else + arrCopy(frame_, _frame); + } + } + + void VideoFrameSource_GPU::reset() + { + reader_.close(); + reader_.open(fileName_); + CV_Assert( reader_.isOpened() ); + } +} + +Ptr cv::superres::createFrameSource_Video_GPU(const string& fileName) +{ + return new VideoFrameSource(fileName); +} + +#endif // HAVE_OPENCV_GPU diff --git a/modules/superres/src/input_array_utility.cpp b/modules/superres/src/input_array_utility.cpp new file mode 100644 index 0000000..8d905bf --- /dev/null +++ b/modules/superres/src/input_array_utility.cpp @@ -0,0 +1,273 @@ +/*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) 2009-2011, 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" + +using namespace std; +using namespace cv; +using namespace cv::gpu; + +Mat cv::superres::arrGetMat(InputArray arr, Mat& buf) +{ + switch (arr.kind()) + { + case _InputArray::GPU_MAT: + arr.getGpuMat().download(buf); + return buf; + + case _InputArray::OPENGL_BUFFER: + arr.getOGlBuffer().copyTo(buf); + return buf; + + case _InputArray::OPENGL_TEXTURE: + arr.getOGlTexture2D().copyTo(buf); + return buf; + + default: + return arr.getMat(); + } +} + +GpuMat cv::superres::arrGetGpuMat(InputArray arr, GpuMat& buf) +{ + switch (arr.kind()) + { + case _InputArray::GPU_MAT: + return arr.getGpuMat(); + + case _InputArray::OPENGL_BUFFER: + arr.getOGlBuffer().copyTo(buf); + return buf; + + case _InputArray::OPENGL_TEXTURE: + arr.getOGlTexture2D().copyTo(buf); + return buf; + + default: + buf.upload(arr.getMat()); + return buf; + } +} + +namespace +{ + void mat2mat(InputArray src, OutputArray dst) + { + src.getMat().copyTo(dst); + } + void arr2buf(InputArray src, OutputArray dst) + { + dst.getOGlBufferRef().copyFrom(src); + } + void arr2tex(InputArray src, OutputArray dst) + { + dst.getOGlTexture2D().copyFrom(src); + } + void mat2gpu(InputArray src, OutputArray dst) + { + dst.getGpuMatRef().upload(src.getMat()); + } + void buf2arr(InputArray src, OutputArray dst) + { + src.getOGlBuffer().copyTo(dst); + } + void tex2arr(InputArray src, OutputArray dst) + { + src.getOGlTexture2D().copyTo(dst); + } + void gpu2mat(InputArray src, OutputArray dst) + { + GpuMat d = src.getGpuMat(); + dst.create(d.size(), d.type()); + Mat m = dst.getMat(); + d.download(m); + } + void gpu2gpu(InputArray src, OutputArray dst) + { + src.getGpuMat().copyTo(dst.getGpuMatRef()); + } +} + +void cv::superres::arrCopy(InputArray src, OutputArray dst) +{ + typedef void (*func_t)(InputArray src, OutputArray dst); + static const func_t funcs[10][10] = + { + {0, 0, 0, 0, 0, 0, 0, 0, 0, 0}, + {0, mat2mat, mat2mat, mat2mat, mat2mat, mat2mat, mat2mat, arr2buf, arr2tex, mat2gpu}, + {0, mat2mat, mat2mat, mat2mat, mat2mat, mat2mat, mat2mat, arr2buf, arr2tex, mat2gpu}, + {0, mat2mat, mat2mat, mat2mat, mat2mat, mat2mat, mat2mat, arr2buf, arr2tex, mat2gpu}, + {0, mat2mat, mat2mat, mat2mat, mat2mat, mat2mat, mat2mat, arr2buf, arr2tex, mat2gpu}, + {0, mat2mat, mat2mat, mat2mat, mat2mat, mat2mat, mat2mat, arr2buf, arr2tex, mat2gpu}, + {0, mat2mat, mat2mat, mat2mat, mat2mat, mat2mat, mat2mat, arr2buf, arr2tex, mat2gpu}, + {0, buf2arr, buf2arr, buf2arr, buf2arr, buf2arr, buf2arr, buf2arr, buf2arr, buf2arr}, + {0, tex2arr, tex2arr, tex2arr, tex2arr, tex2arr, tex2arr, tex2arr, tex2arr, tex2arr}, + {0, gpu2mat, gpu2mat, gpu2mat, gpu2mat, gpu2mat, gpu2mat, arr2buf, arr2tex, gpu2gpu} + }; + + const int src_kind = src.kind() >> _InputArray::KIND_SHIFT; + const int dst_kind = dst.kind() >> _InputArray::KIND_SHIFT; + + CV_DbgAssert( src_kind >= 0 && src_kind < 10 ); + CV_DbgAssert( dst_kind >= 0 && dst_kind < 10 ); + + const func_t func = funcs[src_kind][dst_kind]; + CV_DbgAssert( func != 0 ); + + func(src, dst); +} + +namespace +{ + void convertToCn(InputArray src, OutputArray dst, int cn) + { + CV_Assert( src.channels() == 1 || src.channels() == 3 || src.channels() == 4 ); + CV_Assert( cn == 1 || cn == 3 || cn == 4 ); + + static const int codes[5][5] = + { + {-1, -1, -1, -1, -1}, + {-1, -1, -1, COLOR_GRAY2BGR, COLOR_GRAY2BGRA}, + {-1, -1, -1, -1, -1}, + {-1, COLOR_BGR2GRAY, -1, -1, COLOR_BGR2BGRA}, + {-1, COLOR_BGRA2GRAY, -1, COLOR_BGRA2BGR, -1}, + }; + + const int code = codes[src.channels()][cn]; + CV_DbgAssert( code >= 0 ); + + switch (src.kind()) + { + case _InputArray::GPU_MAT: + #ifdef HAVE_OPENCV_GPU + gpu::cvtColor(src.getGpuMat(), dst.getGpuMatRef(), code, cn); + #else + CV_Error(CV_StsNotImplemented, "The called functionality is disabled for current build or platform"); + #endif + break; + + default: + cvtColor(src, dst, code, cn); + break; + } + } + + void convertToDepth(InputArray src, OutputArray dst, int depth) + { + CV_Assert( src.depth() <= CV_64F ); + CV_Assert( depth == CV_8U || depth == CV_32F ); + + static const double maxVals[] = + { + numeric_limits::max(), + numeric_limits::max(), + numeric_limits::max(), + numeric_limits::max(), + numeric_limits::max(), + 1.0, + 1.0, + }; + + const double scale = maxVals[depth] / maxVals[src.depth()]; + + switch (src.kind()) + { + case _InputArray::GPU_MAT: + src.getGpuMat().convertTo(dst.getGpuMatRef(), depth, scale); + break; + + default: + src.getMat().convertTo(dst, depth, scale); + break; + } + } +} + +Mat cv::superres::convertToType(const Mat& src, int type, Mat& buf0, Mat& buf1) +{ + if (src.type() == type) + return src; + + const int depth = CV_MAT_DEPTH(type); + const int cn = CV_MAT_CN(type); + + if (src.depth() == depth) + { + convertToCn(src, buf0, cn); + return buf0; + } + + if (src.channels() == cn) + { + convertToDepth(src, buf1, depth); + return buf1; + } + + convertToCn(src, buf0, cn); + convertToDepth(buf0, buf1, depth); + return buf1; +} + +GpuMat cv::superres::convertToType(const GpuMat& src, int type, GpuMat& buf0, GpuMat& buf1) +{ + if (src.type() == type) + return src; + + const int depth = CV_MAT_DEPTH(type); + const int cn = CV_MAT_CN(type); + + if (src.depth() == depth) + { + convertToCn(src, buf0, cn); + return buf0; + } + + if (src.channels() == cn) + { + convertToDepth(src, buf1, depth); + return buf1; + } + + convertToCn(src, buf0, cn); + convertToDepth(buf0, buf1, depth); + return buf1; +} diff --git a/modules/superres/src/input_array_utility.hpp b/modules/superres/src/input_array_utility.hpp new file mode 100644 index 0000000..790d621 --- /dev/null +++ b/modules/superres/src/input_array_utility.hpp @@ -0,0 +1,63 @@ +/*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) 2009-2011, 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*/ + +#ifndef __OPENCV_SUPERRES_INPUT_ARRAY_UTILITY_HPP__ +#define __OPENCV_SUPERRES_INPUT_ARRAY_UTILITY_HPP__ + +#include "opencv2/core/core.hpp" +#include "opencv2/core/gpumat.hpp" + +namespace cv +{ + namespace superres + { + CV_EXPORTS Mat arrGetMat(InputArray arr, Mat& buf); + CV_EXPORTS gpu::GpuMat arrGetGpuMat(InputArray arr, gpu::GpuMat& buf); + + CV_EXPORTS void arrCopy(InputArray src, OutputArray dst); + + CV_EXPORTS Mat convertToType(const Mat& src, int type, Mat& buf0, Mat& buf1); + CV_EXPORTS gpu::GpuMat convertToType(const gpu::GpuMat& src, int type, gpu::GpuMat& buf0, gpu::GpuMat& buf1); + } +} + +#endif // __OPENCV_SUPERRES_INPUT_ARRAY_UTILITY_HPP__ diff --git a/modules/superres/src/optical_flow.cpp b/modules/superres/src/optical_flow.cpp new file mode 100644 index 0000000..8c8454c --- /dev/null +++ b/modules/superres/src/optical_flow.cpp @@ -0,0 +1,721 @@ +/*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) 2009-2011, 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" + +using namespace std; +using namespace cv; +using namespace cv::gpu; +using namespace cv::superres; +using namespace cv::superres::detail; + +/////////////////////////////////////////////////////////////////// +// CpuOpticalFlow + +namespace +{ + class CpuOpticalFlow : public DenseOpticalFlowExt + { + public: + explicit CpuOpticalFlow(int work_type); + + void calc(InputArray frame0, InputArray frame1, OutputArray flow1, OutputArray flow2); + void collectGarbage(); + + protected: + virtual void impl(const Mat& input0, const Mat& input1, OutputArray dst) = 0; + + private: + int work_type_; + Mat buf_[6]; + Mat flow_; + Mat flows_[2]; + }; + + CpuOpticalFlow::CpuOpticalFlow(int work_type) : work_type_(work_type) + { + } + + void CpuOpticalFlow::calc(InputArray _frame0, InputArray _frame1, OutputArray _flow1, OutputArray _flow2) + { + Mat frame0 = arrGetMat(_frame0, buf_[0]); + Mat frame1 = arrGetMat(_frame1, buf_[1]); + + CV_Assert( frame1.type() == frame0.type() ); + CV_Assert( frame1.size() == frame0.size() ); + + Mat input0 = convertToType(frame0, work_type_, buf_[2], buf_[3]); + Mat input1 = convertToType(frame1, work_type_, buf_[4], buf_[5]); + + if (!_flow2.needed() && _flow1.kind() < _InputArray::OPENGL_BUFFER) + { + impl(input0, input1, _flow1); + return; + } + + impl(input0, input1, flow_); + + if (!_flow2.needed()) + { + arrCopy(flow_, _flow1); + } + else + { + split(flow_, flows_); + + arrCopy(flows_[0], _flow1); + arrCopy(flows_[1], _flow2); + } + } + + void CpuOpticalFlow::collectGarbage() + { + for (int i = 0; i < 6; ++i) + buf_[i].release(); + flow_.release(); + flows_[0].release(); + flows_[1].release(); + } +} + +/////////////////////////////////////////////////////////////////// +// Farneback + +namespace +{ + class Farneback : public CpuOpticalFlow + { + public: + AlgorithmInfo* info() const; + + Farneback(); + + protected: + void impl(const Mat& input0, const Mat& input1, OutputArray dst); + + private: + double pyrScale_; + int numLevels_; + int winSize_; + int numIters_; + int polyN_; + double polySigma_; + int flags_; + }; + + CV_INIT_ALGORITHM(Farneback, "DenseOpticalFlowExt.Farneback", + obj.info()->addParam(obj, "pyrScale", obj.pyrScale_); + obj.info()->addParam(obj, "numLevels", obj.numLevels_); + obj.info()->addParam(obj, "winSize", obj.winSize_); + obj.info()->addParam(obj, "numIters", obj.numIters_); + obj.info()->addParam(obj, "polyN", obj.polyN_); + obj.info()->addParam(obj, "polySigma", obj.polySigma_); + obj.info()->addParam(obj, "flags", obj.flags_)); + + Farneback::Farneback() : CpuOpticalFlow(CV_8UC1) + { + pyrScale_ = 0.5; + numLevels_ = 5; + winSize_ = 13; + numIters_ = 10; + polyN_ = 5; + polySigma_ = 1.1; + flags_ = 0; + } + + void Farneback::impl(const Mat& input0, const Mat& input1, OutputArray dst) + { + calcOpticalFlowFarneback(input0, input1, dst, pyrScale_, numLevels_, winSize_, numIters_, polyN_, polySigma_, flags_); + } +} + +Ptr cv::superres::createOptFlow_Farneback() +{ + return new Farneback; +} + +/////////////////////////////////////////////////////////////////// +// Simple + +namespace +{ + class Simple : public CpuOpticalFlow + { + public: + AlgorithmInfo* info() const; + + Simple(); + + protected: + void impl(const Mat& input0, const Mat& input1, OutputArray dst); + + private: + int layers_; + int averagingBlockSize_; + int maxFlow_; + double sigmaDist_; + double sigmaColor_; + int postProcessWindow_; + double sigmaDistFix_; + double sigmaColorFix_; + double occThr_; + int upscaleAveragingRadius_; + double upscaleSigmaDist_; + double upscaleSigmaColor_; + double speedUpThr_; + }; + + CV_INIT_ALGORITHM(Simple, "DenseOpticalFlowExt.Simple", + obj.info()->addParam(obj, "layers", obj.layers_); + obj.info()->addParam(obj, "averagingBlockSize", obj.averagingBlockSize_); + obj.info()->addParam(obj, "maxFlow", obj.maxFlow_); + obj.info()->addParam(obj, "sigmaDist", obj.sigmaDist_); + obj.info()->addParam(obj, "sigmaColor", obj.sigmaColor_); + obj.info()->addParam(obj, "postProcessWindow", obj.postProcessWindow_); + obj.info()->addParam(obj, "sigmaDistFix", obj.sigmaDistFix_); + obj.info()->addParam(obj, "sigmaColorFix", obj.sigmaColorFix_); + obj.info()->addParam(obj, "occThr", obj.occThr_); + obj.info()->addParam(obj, "upscaleAveragingRadius", obj.upscaleAveragingRadius_); + obj.info()->addParam(obj, "upscaleSigmaDist", obj.upscaleSigmaDist_); + obj.info()->addParam(obj, "upscaleSigmaColor", obj.upscaleSigmaColor_); + obj.info()->addParam(obj, "speedUpThr", obj.speedUpThr_)); + + Simple::Simple() : CpuOpticalFlow(CV_8UC3) + { + layers_ = 3; + averagingBlockSize_ = 2; + maxFlow_ = 4; + sigmaDist_ = 4.1; + sigmaColor_ = 25.5; + postProcessWindow_ = 18; + sigmaDistFix_ = 55.0; + sigmaColorFix_ = 25.5; + occThr_ = 0.35; + upscaleAveragingRadius_ = 18; + upscaleSigmaDist_ = 55.0; + upscaleSigmaColor_ = 25.5; + speedUpThr_ = 10; + } + + void Simple::impl(const Mat& _input0, const Mat& _input1, OutputArray dst) + { + Mat input0 = _input0; + Mat input1 = _input1; + calcOpticalFlowSF(input0, input1, dst.getMatRef(), + layers_, + averagingBlockSize_, + maxFlow_, + sigmaDist_, + sigmaColor_, + postProcessWindow_, + sigmaDistFix_, + sigmaColorFix_, + occThr_, + upscaleAveragingRadius_, + upscaleSigmaDist_, + upscaleSigmaColor_, + speedUpThr_); + } +} + +Ptr cv::superres::createOptFlow_Simple() +{ + return new Simple; +} + +/////////////////////////////////////////////////////////////////// +// DualTVL1 + +namespace +{ + class DualTVL1 : public CpuOpticalFlow + { + public: + AlgorithmInfo* info() const; + + DualTVL1(); + + void collectGarbage(); + + protected: + void impl(const Mat& input0, const Mat& input1, OutputArray dst); + + private: + double tau_; + double lambda_; + double theta_; + int nscales_; + int warps_; + double epsilon_; + int iterations_; + bool useInitialFlow_; + + Ptr alg_; + }; + + CV_INIT_ALGORITHM(DualTVL1, "DenseOpticalFlowExt.DualTVL1", + obj.info()->addParam(obj, "tau", obj.tau_); + obj.info()->addParam(obj, "lambda", obj.lambda_); + obj.info()->addParam(obj, "theta", obj.theta_); + obj.info()->addParam(obj, "nscales", obj.nscales_); + obj.info()->addParam(obj, "warps", obj.warps_); + obj.info()->addParam(obj, "epsilon", obj.epsilon_); + obj.info()->addParam(obj, "iterations", obj.iterations_); + obj.info()->addParam(obj, "useInitialFlow", obj.useInitialFlow_)); + + DualTVL1::DualTVL1() : CpuOpticalFlow(CV_8UC1) + { + alg_ = cv::createOptFlow_DualTVL1(); + tau_ = alg_->getDouble("tau"); + lambda_ = alg_->getDouble("lambda"); + theta_ = alg_->getDouble("theta"); + nscales_ = alg_->getInt("nscales"); + warps_ = alg_->getInt("warps"); + epsilon_ = alg_->getDouble("epsilon"); + iterations_ = alg_->getInt("iterations"); + useInitialFlow_ = alg_->getBool("useInitialFlow"); + } + + void DualTVL1::impl(const Mat& input0, const Mat& input1, OutputArray dst) + { + alg_->set("tau", tau_); + alg_->set("lambda", lambda_); + alg_->set("theta", theta_); + alg_->set("nscales", nscales_); + alg_->set("warps", warps_); + alg_->set("epsilon", epsilon_); + alg_->set("iterations", iterations_); + alg_->set("useInitialFlow", useInitialFlow_); + + alg_->calc(input0, input1, dst); + } + + void DualTVL1::collectGarbage() + { + alg_->collectGarbage(); + CpuOpticalFlow::collectGarbage(); + } +} + +Ptr cv::superres::createOptFlow_DualTVL1() +{ + return new DualTVL1; +} + +/////////////////////////////////////////////////////////////////// +// GpuOpticalFlow + +#ifndef HAVE_OPENCV_GPU + +Ptr cv::superres::createOptFlow_Farneback_GPU() +{ + CV_Error(CV_StsNotImplemented, "The called functionality is disabled for current build or platform"); + return Ptr(); +} + +Ptr cv::superres::createOptFlow_DualTVL1_GPU() +{ + CV_Error(CV_StsNotImplemented, "The called functionality is disabled for current build or platform"); + return Ptr(); +} + +Ptr cv::superres::createOptFlow_Brox_GPU() +{ + CV_Error(CV_StsNotImplemented, "The called functionality is disabled for current build or platform"); + return Ptr(); +} + +Ptr cv::superres::createOptFlow_PyrLK_GPU() +{ + CV_Error(CV_StsNotImplemented, "The called functionality is disabled for current build or platform"); + return Ptr(); +} + +#else // HAVE_OPENCV_GPU + +namespace +{ + class GpuOpticalFlow : public DenseOpticalFlowExt + { + public: + explicit GpuOpticalFlow(int work_type); + + void calc(InputArray frame0, InputArray frame1, OutputArray flow1, OutputArray flow2); + void collectGarbage(); + + protected: + virtual void impl(const GpuMat& input0, const GpuMat& input1, GpuMat& dst1, GpuMat& dst2) = 0; + + private: + int work_type_; + GpuMat buf_[6]; + GpuMat u_, v_, flow_; + }; + + GpuOpticalFlow::GpuOpticalFlow(int work_type) : work_type_(work_type) + { + } + + void GpuOpticalFlow::calc(InputArray _frame0, InputArray _frame1, OutputArray _flow1, OutputArray _flow2) + { + GpuMat frame0 = arrGetGpuMat(_frame0, buf_[0]); + GpuMat frame1 = arrGetGpuMat(_frame1, buf_[1]); + + CV_Assert( frame1.type() == frame0.type() ); + CV_Assert( frame1.size() == frame0.size() ); + + GpuMat input0 = convertToType(frame0, work_type_, buf_[2], buf_[3]); + GpuMat input1 = convertToType(frame1, work_type_, buf_[4], buf_[5]); + + if (_flow2.needed() && _flow1.kind() == _InputArray::GPU_MAT && _flow2.kind() == _InputArray::GPU_MAT) + { + impl(input0, input1, _flow1.getGpuMatRef(), _flow2.getGpuMatRef()); + return; + } + + impl(input0, input1, u_, v_); + + if (_flow2.needed()) + { + arrCopy(u_, _flow1); + arrCopy(v_, _flow2); + } + else + { + GpuMat src[] = {u_, v_}; + merge(src, 2, flow_); + arrCopy(flow_, _flow1); + } + } + + void GpuOpticalFlow::collectGarbage() + { + for (int i = 0; i < 6; ++i) + buf_[i].release(); + u_.release(); + v_.release(); + flow_.release(); + } +} + +/////////////////////////////////////////////////////////////////// +// Brox_GPU + +namespace +{ + class Brox_GPU : public GpuOpticalFlow + { + public: + AlgorithmInfo* info() const; + + Brox_GPU(); + + void collectGarbage(); + + protected: + void impl(const GpuMat& input0, const GpuMat& input1, GpuMat& dst1, GpuMat& dst2); + + private: + double alpha_; + double gamma_; + double scaleFactor_; + int innerIterations_; + int outerIterations_; + int solverIterations_; + + BroxOpticalFlow alg_; + }; + + CV_INIT_ALGORITHM(Brox_GPU, "DenseOpticalFlowExt.Brox_GPU", + obj.info()->addParam(obj, "alpha", obj.alpha_, false, 0, 0, "Flow smoothness"); + obj.info()->addParam(obj, "gamma", obj.gamma_, false, 0, 0, "Gradient constancy importance"); + obj.info()->addParam(obj, "scaleFactor", obj.scaleFactor_, false, 0, 0, "Pyramid scale factor"); + obj.info()->addParam(obj, "innerIterations", obj.innerIterations_, false, 0, 0, "Number of lagged non-linearity iterations (inner loop)"); + obj.info()->addParam(obj, "outerIterations", obj.outerIterations_, false, 0, 0, "Number of warping iterations (number of pyramid levels)"); + obj.info()->addParam(obj, "solverIterations", obj.solverIterations_, false, 0, 0, "Number of linear system solver iterations")); + + Brox_GPU::Brox_GPU() : GpuOpticalFlow(CV_32FC1), alg_(0.197f, 50.0f, 0.8f, 10, 77, 10) + { + alpha_ = alg_.alpha; + gamma_ = alg_.gamma; + scaleFactor_ = alg_.scale_factor; + innerIterations_ = alg_.inner_iterations; + outerIterations_ = alg_.outer_iterations; + solverIterations_ = alg_.solver_iterations; + } + + void Brox_GPU::impl(const GpuMat& input0, const GpuMat& input1, GpuMat& dst1, GpuMat& dst2) + { + alg_.alpha = static_cast(alpha_); + alg_.gamma = static_cast(gamma_); + alg_.scale_factor = static_cast(scaleFactor_); + alg_.inner_iterations = innerIterations_; + alg_.outer_iterations = outerIterations_; + alg_.solver_iterations = solverIterations_; + + alg_(input0, input1, dst1, dst2); + } + + void Brox_GPU::collectGarbage() + { + alg_.buf.release(); + GpuOpticalFlow::collectGarbage(); + } +} + +Ptr cv::superres::createOptFlow_Brox_GPU() +{ + return new Brox_GPU; +} + +/////////////////////////////////////////////////////////////////// +// PyrLK_GPU + +namespace +{ + class PyrLK_GPU : public GpuOpticalFlow + { + public: + AlgorithmInfo* info() const; + + PyrLK_GPU(); + + void collectGarbage(); + + protected: + void impl(const GpuMat& input0, const GpuMat& input1, GpuMat& dst1, GpuMat& dst2); + + private: + int winSize_; + int maxLevel_; + int iterations_; + + PyrLKOpticalFlow alg_; + }; + + CV_INIT_ALGORITHM(PyrLK_GPU, "DenseOpticalFlowExt.PyrLK_GPU", + obj.info()->addParam(obj, "winSize", obj.winSize_); + obj.info()->addParam(obj, "maxLevel", obj.maxLevel_); + obj.info()->addParam(obj, "iterations", obj.iterations_)); + + PyrLK_GPU::PyrLK_GPU() : GpuOpticalFlow(CV_8UC1) + { + winSize_ = alg_.winSize.width; + maxLevel_ = alg_.maxLevel; + iterations_ = alg_.iters; + } + + void PyrLK_GPU::impl(const GpuMat& input0, const GpuMat& input1, GpuMat& dst1, GpuMat& dst2) + { + alg_.winSize.width = winSize_; + alg_.winSize.height = winSize_; + alg_.maxLevel = maxLevel_; + alg_.iters = iterations_; + + alg_.dense(input0, input1, dst1, dst2); + } + + void PyrLK_GPU::collectGarbage() + { + alg_.releaseMemory(); + GpuOpticalFlow::collectGarbage(); + } +} + +Ptr cv::superres::createOptFlow_PyrLK_GPU() +{ + return new PyrLK_GPU; +} + +/////////////////////////////////////////////////////////////////// +// Farneback_GPU + +namespace +{ + class Farneback_GPU : public GpuOpticalFlow + { + public: + AlgorithmInfo* info() const; + + Farneback_GPU(); + + void collectGarbage(); + + protected: + void impl(const GpuMat& input0, const GpuMat& input1, GpuMat& dst1, GpuMat& dst2); + + private: + double pyrScale_; + int numLevels_; + int winSize_; + int numIters_; + int polyN_; + double polySigma_; + int flags_; + + FarnebackOpticalFlow alg_; + }; + + CV_INIT_ALGORITHM(Farneback_GPU, "DenseOpticalFlowExt.Farneback_GPU", + obj.info()->addParam(obj, "pyrScale", obj.pyrScale_); + obj.info()->addParam(obj, "numLevels", obj.numLevels_); + obj.info()->addParam(obj, "winSize", obj.winSize_); + obj.info()->addParam(obj, "numIters", obj.numIters_); + obj.info()->addParam(obj, "polyN", obj.polyN_); + obj.info()->addParam(obj, "polySigma", obj.polySigma_); + obj.info()->addParam(obj, "flags", obj.flags_)); + + Farneback_GPU::Farneback_GPU() : GpuOpticalFlow(CV_8UC1) + { + pyrScale_ = alg_.pyrScale; + numLevels_ = alg_.numLevels; + winSize_ = alg_.winSize; + numIters_ = alg_.numIters; + polyN_ = alg_.polyN; + polySigma_ = alg_.polySigma; + flags_ = alg_.flags; + } + + void Farneback_GPU::impl(const GpuMat& input0, const GpuMat& input1, GpuMat& dst1, GpuMat& dst2) + { + alg_.pyrScale = pyrScale_; + alg_.numLevels = numLevels_; + alg_.winSize = winSize_; + alg_.numIters = numIters_; + alg_.polyN = polyN_; + alg_.polySigma = polySigma_; + alg_.flags = flags_; + + alg_(input0, input1, dst1, dst2); + } + + void Farneback_GPU::collectGarbage() + { + alg_.releaseMemory(); + GpuOpticalFlow::collectGarbage(); + } +} + +Ptr cv::superres::createOptFlow_Farneback_GPU() +{ + return new Farneback_GPU; +} + +/////////////////////////////////////////////////////////////////// +// DualTVL1_GPU + +namespace +{ + class DualTVL1_GPU : public GpuOpticalFlow + { + public: + AlgorithmInfo* info() const; + + DualTVL1_GPU(); + + void collectGarbage(); + + protected: + void impl(const GpuMat& input0, const GpuMat& input1, GpuMat& dst1, GpuMat& dst2); + + private: + double tau_; + double lambda_; + double theta_; + int nscales_; + int warps_; + double epsilon_; + int iterations_; + bool useInitialFlow_; + + OpticalFlowDual_TVL1_GPU alg_; + }; + + CV_INIT_ALGORITHM(DualTVL1_GPU, "DenseOpticalFlowExt.DualTVL1_GPU", + obj.info()->addParam(obj, "tau", obj.tau_); + obj.info()->addParam(obj, "lambda", obj.lambda_); + obj.info()->addParam(obj, "theta", obj.theta_); + obj.info()->addParam(obj, "nscales", obj.nscales_); + obj.info()->addParam(obj, "warps", obj.warps_); + obj.info()->addParam(obj, "epsilon", obj.epsilon_); + obj.info()->addParam(obj, "iterations", obj.iterations_); + obj.info()->addParam(obj, "useInitialFlow", obj.useInitialFlow_)); + + DualTVL1_GPU::DualTVL1_GPU() : GpuOpticalFlow(CV_8UC1) + { + tau_ = alg_.tau; + lambda_ = alg_.lambda; + theta_ = alg_.theta; + nscales_ = alg_.nscales; + warps_ = alg_.warps; + epsilon_ = alg_.epsilon; + iterations_ = alg_.iterations; + useInitialFlow_ = alg_.useInitialFlow; + } + + void DualTVL1_GPU::impl(const GpuMat& input0, const GpuMat& input1, GpuMat& dst1, GpuMat& dst2) + { + alg_.tau = tau_; + alg_.lambda = lambda_; + alg_.theta = theta_; + alg_.nscales = nscales_; + alg_.warps = warps_; + alg_.epsilon = epsilon_; + alg_.iterations = iterations_; + alg_.useInitialFlow = useInitialFlow_; + + alg_(input0, input1, dst1, dst2); + } + + void DualTVL1_GPU::collectGarbage() + { + alg_.collectGarbage(); + GpuOpticalFlow::collectGarbage(); + } +} + +Ptr cv::superres::createOptFlow_DualTVL1_GPU() +{ + return new DualTVL1_GPU; +} + +#endif // HAVE_OPENCV_GPU diff --git a/modules/superres/src/precomp.cpp b/modules/superres/src/precomp.cpp new file mode 100644 index 0000000..1113852 --- /dev/null +++ b/modules/superres/src/precomp.cpp @@ -0,0 +1,43 @@ +/*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) 2009-2011, 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" diff --git a/modules/superres/src/precomp.hpp b/modules/superres/src/precomp.hpp new file mode 100644 index 0000000..2b4d0d9 --- /dev/null +++ b/modules/superres/src/precomp.hpp @@ -0,0 +1,78 @@ +/*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) 2009-2011, 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*/ + +#ifndef __OPENCV_PRECOMP_H__ +#define __OPENCV_PRECOMP_H__ + +#include +#include + +#ifdef HAVE_CVCONFIG_H + #include "cvconfig.h" +#endif + +#include "opencv2/opencv_modules.hpp" +#include "opencv2/core/core.hpp" +#include "opencv2/core/gpumat.hpp" +#include "opencv2/core/opengl_interop.hpp" +#include "opencv2/core/internal.hpp" +#include "opencv2/imgproc/imgproc.hpp" +#include "opencv2/video/tracking.hpp" + +#ifdef HAVE_OPENCV_GPU + #include "opencv2/gpu/gpu.hpp" + #ifdef HAVE_CUDA + #include "opencv2/gpu/stream_accessor.hpp" + #endif +#endif + +#ifdef HAVE_OPENCV_HIGHGUI + #include "opencv2/highgui/highgui.hpp" +#endif + +#include "opencv2/superres/superres.hpp" +#include "opencv2/superres/optical_flow.hpp" +#include "input_array_utility.hpp" + +#include "ring_buffer.hpp" + +#endif /* __OPENCV_PRECOMP_H__ */ diff --git a/modules/superres/src/ring_buffer.hpp b/modules/superres/src/ring_buffer.hpp new file mode 100644 index 0000000..3c51d7a --- /dev/null +++ b/modules/superres/src/ring_buffer.hpp @@ -0,0 +1,79 @@ +/*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) 2009-2011, 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*/ + +#ifndef __RING_BUFFER_HPP__ +#define __RING_BUFFER_HPP__ + +#include "precomp.hpp" + +namespace cv +{ + namespace superres + { + namespace detail + { + template + inline const T& at(int index, const std::vector& items) + { + const int len = static_cast(items.size()); + if (index < 0) + index -= ((index - len + 1) / len) * len; + if (index >= len) + index %= len; + return items[index]; + } + + template + inline T& at(int index, std::vector& items) + { + const int len = static_cast(items.size()); + if (index < 0) + index -= ((index - len + 1) / len) * len; + if (index >= len) + index %= len; + return items[index]; + } + } + } +} + +#endif // __RING_BUFFER_HPP__ diff --git a/modules/superres/src/super_resolution.cpp b/modules/superres/src/super_resolution.cpp new file mode 100644 index 0000000..73a2147 --- /dev/null +++ b/modules/superres/src/super_resolution.cpp @@ -0,0 +1,85 @@ +/*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) 2009-2011, 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" + +using namespace std; +using namespace cv; +using namespace cv::superres; + +bool cv::superres::initModule_superres() +{ + return !createSuperResolution_BTVL1().empty(); +} + +cv::superres::SuperResolution::SuperResolution() +{ + frameSource_ = createFrameSource_Empty(); + firstCall_ = true; +} + +void cv::superres::SuperResolution::setInput(const Ptr& frameSource) +{ + frameSource_ = frameSource; + firstCall_ = true; +} + +void cv::superres::SuperResolution::nextFrame(OutputArray frame) +{ + if (firstCall_) + { + initImpl(frameSource_); + firstCall_ = false; + } + + processImpl(frameSource_, frame); +} + +void cv::superres::SuperResolution::reset() +{ + frameSource_->reset(); + firstCall_ = true; +} + +void cv::superres::SuperResolution::collectGarbage() +{ +} diff --git a/modules/superres/test/test_main.cpp b/modules/superres/test/test_main.cpp new file mode 100644 index 0000000..146e262 --- /dev/null +++ b/modules/superres/test/test_main.cpp @@ -0,0 +1,3 @@ +#include "test_precomp.hpp" + +CV_TEST_MAIN("superres") diff --git a/modules/superres/test/test_precomp.cpp b/modules/superres/test/test_precomp.cpp new file mode 100644 index 0000000..5956e13 --- /dev/null +++ b/modules/superres/test/test_precomp.cpp @@ -0,0 +1 @@ +#include "test_precomp.hpp" diff --git a/modules/superres/test/test_precomp.hpp b/modules/superres/test/test_precomp.hpp new file mode 100644 index 0000000..84c0a76 --- /dev/null +++ b/modules/superres/test/test_precomp.hpp @@ -0,0 +1,23 @@ +#ifdef __GNUC__ +# pragma GCC diagnostic ignored "-Wmissing-declarations" +# if defined __clang__ || defined __APPLE__ +# pragma GCC diagnostic ignored "-Wmissing-prototypes" +# pragma GCC diagnostic ignored "-Wextra" +# endif +#endif + +#ifndef __OPENCV_TEST_PRECOMP_HPP__ +#define __OPENCV_TEST_PRECOMP_HPP__ + +#ifdef HAVE_CVCONFIG_H +#include "cvconfig.h" +#endif + +#include "opencv2/opencv_modules.hpp" +#include "opencv2/core/core.hpp" +#include "opencv2/imgproc/imgproc.hpp" +#include "opencv2/ts/ts.hpp" +#include "opencv2/superres/superres.hpp" +#include "input_array_utility.hpp" + +#endif diff --git a/modules/superres/test/test_superres.cpp b/modules/superres/test/test_superres.cpp new file mode 100644 index 0000000..44d6a7a --- /dev/null +++ b/modules/superres/test/test_superres.cpp @@ -0,0 +1,236 @@ +#include "test_precomp.hpp" + +class AllignedFrameSource : public cv::superres::FrameSource +{ +public: + AllignedFrameSource(const cv::Ptr& base, int scale); + + void nextFrame(cv::OutputArray frame); + void reset(); + +private: + cv::Ptr base_; + cv::Mat origFrame_; + int scale_; +}; + +AllignedFrameSource::AllignedFrameSource(const cv::Ptr& base, int scale) : + base_(base), scale_(scale) +{ + CV_Assert( !base_.empty() ); +} + +void AllignedFrameSource::nextFrame(cv::OutputArray frame) +{ + base_->nextFrame(origFrame_); + + if (origFrame_.rows % scale_ == 0 && origFrame_.cols % scale_ == 0) + { + cv::superres::arrCopy(origFrame_, frame); + } + else + { + cv::Rect ROI(0, 0, (origFrame_.cols / scale_) * scale_, (origFrame_.rows / scale_) * scale_); + cv::superres::arrCopy(origFrame_(ROI), frame); + } +} + +void AllignedFrameSource::reset() +{ + base_->reset(); +} + +class DegradeFrameSource : public cv::superres::FrameSource +{ +public: + DegradeFrameSource(const cv::Ptr& base, int scale); + + void nextFrame(cv::OutputArray frame); + void reset(); + +private: + cv::Ptr base_; + cv::Mat origFrame_; + cv::Mat blurred_; + cv::Mat deg_; + double iscale_; +}; + +DegradeFrameSource::DegradeFrameSource(const cv::Ptr& base, int scale) : + base_(base), iscale_(1.0 / scale) +{ + CV_Assert( !base_.empty() ); +} + +void addGaussNoise(cv::Mat& image, double sigma) +{ + cv::Mat noise(image.size(), CV_32FC(image.channels())); + cvtest::TS::ptr()->get_rng().fill(noise, cv::RNG::NORMAL, 0.0, sigma); + + cv::addWeighted(image, 1.0, noise, 1.0, 0.0, image, image.depth()); +} + +void addSpikeNoise(cv::Mat& image, int frequency) +{ + cv::Mat_ mask(image.size(), 0); + + for (int y = 0; y < mask.rows; ++y) + { + for (int x = 0; x < mask.cols; ++x) + { + if (cvtest::TS::ptr()->get_rng().uniform(0, frequency) < 1) + mask(y, x) = 255; + } + } + + image.setTo(cv::Scalar::all(255), mask); +} + +void DegradeFrameSource::nextFrame(cv::OutputArray frame) +{ + base_->nextFrame(origFrame_); + + cv::GaussianBlur(origFrame_, blurred_, cv::Size(5, 5), 0); + cv::resize(blurred_, deg_, cv::Size(), iscale_, iscale_, cv::INTER_NEAREST); + + addGaussNoise(deg_, 10.0); + addSpikeNoise(deg_, 500); + + cv::superres::arrCopy(deg_, frame); +} + +void DegradeFrameSource::reset() +{ + base_->reset(); +} + +double MSSIM(const cv::Mat& i1, const cv::Mat& i2) +{ + const double C1 = 6.5025; + const double C2 = 58.5225; + + const int depth = CV_32F; + + cv::Mat I1, I2; + i1.convertTo(I1, depth); + i2.convertTo(I2, depth); + + cv::Mat I2_2 = I2.mul(I2); // I2^2 + cv::Mat I1_2 = I1.mul(I1); // I1^2 + cv::Mat I1_I2 = I1.mul(I2); // I1 * I2 + + cv::Mat mu1, mu2; + cv::GaussianBlur(I1, mu1, cv::Size(11, 11), 1.5); + cv::GaussianBlur(I2, mu2, cv::Size(11, 11), 1.5); + + cv::Mat mu1_2 = mu1.mul(mu1); + cv::Mat mu2_2 = mu2.mul(mu2); + cv::Mat mu1_mu2 = mu1.mul(mu2); + + cv::Mat sigma1_2, sigma2_2, sigma12; + + cv::GaussianBlur(I1_2, sigma1_2, cv::Size(11, 11), 1.5); + sigma1_2 -= mu1_2; + + cv::GaussianBlur(I2_2, sigma2_2, cv::Size(11, 11), 1.5); + sigma2_2 -= mu2_2; + + cv::GaussianBlur(I1_I2, sigma12, cv::Size(11, 11), 1.5); + sigma12 -= mu1_mu2; + + cv::Mat t1, t2; + cv::Mat numerator; + cv::Mat denominator; + + // t3 = ((2*mu1_mu2 + C1).*(2*sigma12 + C2)) + t1 = 2 * mu1_mu2 + C1; + t2 = 2 * sigma12 + C2; + numerator = t1.mul(t2); + + // t1 =((mu1_2 + mu2_2 + C1).*(sigma1_2 + sigma2_2 + C2)) + t1 = mu1_2 + mu2_2 + C1; + t2 = sigma1_2 + sigma2_2 + C2; + denominator = t1.mul(t2); + + // ssim_map = numerator./denominator; + cv::Mat ssim_map; + cv::divide(numerator, denominator, ssim_map); + + // mssim = average of ssim map + cv::Scalar mssim = cv::mean(ssim_map); + + if (i1.channels() == 1) + return mssim[0]; + + return (mssim[0] + mssim[1] + mssim[3]) / 3; +} + +class SuperResolution : public testing::Test +{ +public: + void RunTest(cv::Ptr superRes); +}; + +void SuperResolution::RunTest(cv::Ptr superRes) +{ + const std::string inputVideoName = cvtest::TS::ptr()->get_data_path() + "car.avi"; + const int scale = 2; + const int iterations = 100; + const int temporalAreaRadius = 2; + + ASSERT_FALSE( superRes.empty() ); + + const int btvKernelSize = superRes->getInt("btvKernelSize"); + + superRes->set("scale", scale); + superRes->set("iterations", iterations); + superRes->set("temporalAreaRadius", temporalAreaRadius); + + cv::Ptr goldSource(new AllignedFrameSource(cv::superres::createFrameSource_Video(inputVideoName), scale)); + cv::Ptr lowResSource(new DegradeFrameSource(new AllignedFrameSource(cv::superres::createFrameSource_Video(inputVideoName), scale), scale)); + + // skip first frame + cv::Mat frame; + + lowResSource->nextFrame(frame); + goldSource->nextFrame(frame); + + cv::Rect inner(btvKernelSize, btvKernelSize, frame.cols - 2 * btvKernelSize, frame.rows - 2 * btvKernelSize); + + superRes->setInput(lowResSource); + + double srAvgMSSIM = 0.0; + const int count = 10; + + cv::Mat goldFrame, superResFrame; + for (int i = 0; i < count; ++i) + { + goldSource->nextFrame(goldFrame); + ASSERT_FALSE( goldFrame.empty() ); + + superRes->nextFrame(superResFrame); + ASSERT_FALSE( superResFrame.empty() ); + + const double srMSSIM = MSSIM(goldFrame(inner), superResFrame); + + srAvgMSSIM += srMSSIM; + } + + srAvgMSSIM /= count; + + EXPECT_GE( srAvgMSSIM, 0.5 ); +} + +TEST_F(SuperResolution, BTVL1) +{ + RunTest(cv::superres::createSuperResolution_BTVL1()); +} + +#if defined(HAVE_OPENCV_GPU) && defined(HAVE_CUDA) + +TEST_F(SuperResolution, BTVL1_GPU) +{ + RunTest(cv::superres::createSuperResolution_BTVL1_GPU()); +} + +#endif diff --git a/samples/gpu/CMakeLists.txt b/samples/gpu/CMakeLists.txt index 6abb7e5..9fd19e2 100644 --- a/samples/gpu/CMakeLists.txt +++ b/samples/gpu/CMakeLists.txt @@ -1,7 +1,7 @@ SET(OPENCV_GPU_SAMPLES_REQUIRED_DEPS opencv_core opencv_flann opencv_imgproc opencv_highgui opencv_ml opencv_video opencv_objdetect opencv_features2d opencv_calib3d opencv_legacy opencv_contrib opencv_gpu - opencv_nonfree) + opencv_nonfree opencv_superres) ocv_check_dependencies(${OPENCV_GPU_SAMPLES_REQUIRED_DEPS}) diff --git a/samples/gpu/super_resolution.cpp b/samples/gpu/super_resolution.cpp new file mode 100644 index 0000000..2dd3656 --- /dev/null +++ b/samples/gpu/super_resolution.cpp @@ -0,0 +1,152 @@ +#include +#include +#include +#include "opencv2/core/core.hpp" +#include "opencv2/highgui/highgui.hpp" +#include "opencv2/imgproc/imgproc.hpp" +#include "opencv2/contrib/contrib.hpp" +#include "opencv2/superres/superres.hpp" +#include "opencv2/superres/optical_flow.hpp" + +using namespace std; +using namespace cv; +using namespace cv::superres; + +#define MEASURE_TIME(op) \ + { \ + TickMeter tm; \ + tm.start(); \ + op; \ + tm.stop(); \ + cout << tm.getTimeSec() << " sec" << endl; \ + } + +static Ptr createOptFlow(const string& name, bool useGpu) +{ + if (name == "farneback") + { + if (useGpu) + return createOptFlow_Farneback_GPU(); + else + return createOptFlow_Farneback(); + } + else if (name == "simple") + return createOptFlow_Simple(); + else if (name == "tvl1") + { + if (useGpu) + return createOptFlow_DualTVL1_GPU(); + else + return createOptFlow_DualTVL1(); + } + else if (name == "brox") + return createOptFlow_Brox_GPU(); + else if (name == "pyrlk") + return createOptFlow_PyrLK_GPU(); + else + { + cerr << "Incorrect Optical Flow algorithm - " << name << endl; + exit(-1); + } + + return Ptr(); +} + +int main(int argc, const char* argv[]) +{ + CommandLineParser cmd(argc, argv, + "{ v | video | | Input video }" + "{ o | output | | Output video }" + "{ s | scale | 4 | Scale factor }" + "{ i | iterations | 180 | Iteration count }" + "{ t | temporal | 4 | Radius of the temporal search area }" + "{ f | flow | farneback | Optical flow algorithm (farneback, simple, tvl1, brox, pyrlk) }" + "{ gpu | gpu | false | Use GPU }" + "{ h | help | false | Print help message }" + ); + + if (cmd.get("help")) + { + cout << "This sample demonstrates Super Resolution algorithms for video sequence" << endl; + cmd.printParams(); + return 0; + } + + const string inputVideoName = cmd.get("video"); + const string outputVideoName = cmd.get("output"); + const int scale = cmd.get("scale"); + const int iterations = cmd.get("iterations"); + const int temporalAreaRadius = cmd.get("temporal"); + const string optFlow = cmd.get("flow"); + const bool useGpu = cmd.get("gpu"); + + Ptr superRes; + if (useGpu) + superRes = createSuperResolution_BTVL1_GPU(); + else + superRes = createSuperResolution_BTVL1(); + + superRes->set("scale", scale); + superRes->set("iterations", iterations); + superRes->set("temporalAreaRadius", temporalAreaRadius); + superRes->set("opticalFlow", createOptFlow(optFlow, useGpu)); + + Ptr frameSource; + if (useGpu) + { + // Try to use gpu Video Decoding + try + { + frameSource = createFrameSource_Video_GPU(inputVideoName); + Mat frame; + frameSource->nextFrame(frame); + } + catch (const cv::Exception&) + { + frameSource.release(); + } + } + if (frameSource.empty()) + frameSource = createFrameSource_Video(inputVideoName); + + // skip first frame, it is usually corrupted + { + Mat frame; + frameSource->nextFrame(frame); + cout << "Input : " << inputVideoName << " " << frame.size() << endl; + cout << "Scale factor : " << scale << endl; + cout << "Iterations : " << iterations << endl; + cout << "Temporal radius : " << temporalAreaRadius << endl; + cout << "Optical Flow : " << optFlow << endl; + cout << "Mode : " << (useGpu ? "GPU" : "CPU") << endl; + } + + superRes->setInput(frameSource); + + VideoWriter writer; + + for (int i = 0;; ++i) + { + cout << '[' << setw(3) << i << "] : "; + + Mat result; + MEASURE_TIME(superRes->nextFrame(result)); + + if (result.empty()) + break; + + imshow("Super Resolution", result); + + if (waitKey(1000) > 0) + break; + + if (!outputVideoName.empty()) + { + if (!writer.isOpened()) + writer.open(outputVideoName, CV_FOURCC('X', 'V', 'I', 'D'), 25.0, result.size()); + writer << result; + } + } + + return 0; +}