Task02 Тернопол Леонид СПбГУ

.gitignore

@@ -1,4 +1,8 @@
 **/*_cl.h
 .idea
-build
+/build*
 cmake-build*
+/.cache/
+/.venv/
+/.vscode
+/data/debug/

src/phg/matching/descriptor_matcher.cpp

@@ -1,27 +1,46 @@
 #include "descriptor_matcher.h"
 
-#include <opencv2/flann/miniflann.hpp>
 #include "flann_factory.h"
+#include <algorithm>
+#include <opencv2/core/types.hpp>
+#include <opencv2/flann/miniflann.hpp>
+#include <set>
+
+ /* @TUNABLE */
+static constexpr float RATIO_THRESHOLD = 0.75;
+static constexpr int SEARCH_CHECKS = 10;
 
-void phg::DescriptorMatcher::filterMatchesRatioTest(const std::vector<std::vector<cv::DMatch>> &matches,
-                                                    std::vector<cv::DMatch> &filtered_matches)
+void phg::DescriptorMatcher::filterMatchesRatioTest(const std::vector<std::vector<cv::DMatch>>& matches, std::vector<cv::DMatch>& filtered_matches)
 {
     filtered_matches.clear();
-
-    throw std::runtime_error("not implemented yet");
+    const cv::DMatch* heap[3];
+    auto cmp = [](const cv::DMatch* a, const cv::DMatch* b) { return a->distance < b->distance; };
+    for (auto& match_row : matches) {
+        int size = 0;
+        for (auto& match : match_row) {
+            heap[size++] = &match;
+            std::push_heap(heap, heap + size, cmp);
+            if (size == 3)
+                std::pop_heap(heap, heap + size--, cmp);
+        }
+        if (size == 2) {
+            std::pop_heap(heap, heap + size, cmp);
+            if (heap[0]->distance / heap[1]->distance < RATIO_THRESHOLD * RATIO_THRESHOLD) {
+                filtered_matches.push_back(*heap[0]);
+            }
+        } else if (size == 1) {
+            filtered_matches.push_back(*heap[0]);
+        }
+    }
 }
 
-
-void phg::DescriptorMatcher::filterMatchesClusters(const std::vector<cv::DMatch> &matches,
-                                                   const std::vector<cv::KeyPoint> keypoints_query,
-                                                   const std::vector<cv::KeyPoint> keypoints_train,
-                                                   std::vector<cv::DMatch> &filtered_matches)
+void phg::DescriptorMatcher::filterMatchesClusters(const std::vector<cv::DMatch>& matches, const std::vector<cv::KeyPoint> keypoints_query, const std::vector<cv::KeyPoint> keypoints_train, std::vector<cv::DMatch>& filtered_matches)
 {
     filtered_matches.clear();
 
-    const size_t  total_neighbours  = 5;  // total number of neighbours to test (including candidate)
-    const size_t  consistent_matches  = 3;  // minimum number of consistent matches (including candidate)
-    const float  radius_limit_scale  = 2.f;  // limit search radius by scaled median
+    const size_t total_neighbours = 5; // total number of neighbours to test (including candidate)
+    const size_t consistent_matches = 3; // minimum number of consistent matches (including candidate)
+    const float radius_limit_scale = 2.f; // limit search radius by scaled median
 
     const int n_matches = matches.size();
 
@@ -35,42 +54,68 @@ void phg::DescriptorMatcher::filterMatchesClusters(const std::vector<cv::DMatch>
         points_query.at<cv::Point2f>(i) = keypoints_query[matches[i].queryIdx].pt;
         points_train.at<cv::Point2f>(i) = keypoints_train[matches[i].trainIdx].pt;
     }
-//
-//    // размерность всего 2, так что точное KD-дерево
-//    std::shared_ptr<cv::flann::IndexParams> index_params = flannKdTreeIndexParams(TODO);
-//    std::shared_ptr<cv::flann::SearchParams> search_params = flannKsTreeSearchParams(TODO);
-//
-//    std::shared_ptr<cv::flann::Index> index_query = flannKdTreeIndex(points_query, index_params);
-//    std::shared_ptr<cv::flann::Index> index_train = flannKdTreeIndex(points_train, index_params);
-//
-//    // для каждой точки найти total neighbors ближайших соседей
-//    cv::Mat indices_query(n_matches, total_neighbours, CV_32SC1);
-//    cv::Mat distances2_query(n_matches, total_neighbours, CV_32FC1);
-//    cv::Mat indices_train(n_matches, total_neighbours, CV_32SC1);
-//    cv::Mat distances2_train(n_matches, total_neighbours, CV_32FC1);
-//
-//    index_query->knnSearch(points_query, indices_query, distances2_query, total_neighbours, *search_params);
-//    index_train->knnSearch(points_train, indices_train, distances2_train, total_neighbours, *search_params);
-//
-//    // оценить радиус поиска для каждой картинки
-//    // NB: radius2_query, radius2_train: квадраты радиуса!
-//    float radius2_query, radius2_train;
-//    {
-//        std::vector<double> max_dists2_query(n_matches);
-//        std::vector<double> max_dists2_train(n_matches);
-//        for (int i = 0; i < n_matches; ++i) {
-//            max_dists2_query[i] = distances2_query.at<float>(i, total_neighbours - 1);
-//            max_dists2_train[i] = distances2_train.at<float>(i, total_neighbours - 1);
-//        }
-//
-//        int median_pos = n_matches / 2;
-//        std::nth_element(max_dists2_query.begin(), max_dists2_query.begin() + median_pos, max_dists2_query.end());
-//        std::nth_element(max_dists2_train.begin(), max_dists2_train.begin() + median_pos, max_dists2_train.end());
-//
-//        radius2_query = max_dists2_query[median_pos] * radius_limit_scale * radius_limit_scale;
-//        radius2_train = max_dists2_train[median_pos] * radius_limit_scale * radius_limit_scale;
-//    }
-//
-//    метч остается, если левое и правое множества первых total_neighbors соседей в радиусах поиска(radius2_query, radius2_train) имеют как минимум consistent_matches общих элементов
-//    // TODO заполнить filtered_matches
+
+    // размерность всего 2, так что точное KD-дерево
+    std::shared_ptr<cv::flann::IndexParams> index_params = flannKdTreeIndexParams(1);
+    std::shared_ptr<cv::flann::SearchParams> search_params = flannKsTreeSearchParams(SEARCH_CHECKS);
+
+    std::shared_ptr<cv::flann::Index> index_query = flannKdTreeIndex(points_query, index_params);
+    std::shared_ptr<cv::flann::Index> index_train = flannKdTreeIndex(points_train, index_params);
+
+    // для каждой точки найти total neighbors ближайших соседей
+    cv::Mat indices_query(n_matches, total_neighbours, CV_32SC1);
+    cv::Mat distances2_query(n_matches, total_neighbours, CV_32FC1);
+    cv::Mat indices_train(n_matches, total_neighbours, CV_32SC1);
+    cv::Mat distances2_train(n_matches, total_neighbours, CV_32FC1);
+
+    index_query->knnSearch(points_query, indices_query, distances2_query, total_neighbours, *search_params);
+    index_train->knnSearch(points_train, indices_train, distances2_train, total_neighbours, *search_params);
+
+    // оценить радиус поиска для каждой картинки
+    // NB: radius2_query, radius2_train: квадраты радиуса!
+    float radius2_query, radius2_train;
+    {
+        std::vector<double> max_dists2_query(n_matches);
+        std::vector<double> max_dists2_train(n_matches);
+        for (int i = 0; i < n_matches; ++i) {
+            max_dists2_query[i] = distances2_query.at<float>(i, total_neighbours - 1);
+            max_dists2_train[i] = distances2_train.at<float>(i, total_neighbours - 1);
+        }
+
+        int median_pos = n_matches / 2;
+        std::nth_element(max_dists2_query.begin(), max_dists2_query.begin() + median_pos, max_dists2_query.end());
+        std::nth_element(max_dists2_train.begin(), max_dists2_train.begin() + median_pos, max_dists2_train.end());
+
+        radius2_query = max_dists2_query[median_pos] * radius_limit_scale * radius_limit_scale;
+        radius2_train = max_dists2_train[median_pos] * radius_limit_scale * radius_limit_scale;
+    }
+
+    //    метч остается, если левое и правое множества первых total_neighbors соседей в радиусах поиска(radius2_query, radius2_train) имеют как минимум consistent_matches общих элементов
+    // заполнить filtered_matches
+
+    for (int i = 0; i < n_matches; ++i) {
+        int n_consistent = 0;
+        for (int src = 0; src < total_neighbours; ++src) {
+            float dsrc = distances2_query.at<float>(i, src);
+            if (dsrc > radius2_query)
+                continue;
+
+            for (int dst = 0; dst < total_neighbours; ++dst) {
+                float ddst = distances2_train.at<float>(i, dst);
+                if (ddst > radius2_train)
+                    continue;
+
+                if (indices_query.at<int>(i, src) == indices_train.at<int>(i, dst)) {
+                    n_consistent += 1;
+
+                    if (n_consistent >= consistent_matches)
+                        goto early_success;
+                }
+            }
+        }
+        if (n_consistent >= consistent_matches) {
+        early_success:
+            filtered_matches.push_back(matches[i]);
+        }
+    }
 }

src/phg/matching/flann_matcher.cpp

@@ -1,13 +1,19 @@
 #include <iostream>
+#include <opencv2/core/hal/interface.h>
+#include <opencv2/core/mat.hpp>
+#include <opencv2/core/types.hpp>
+#include <vector>
 #include "flann_matcher.h"
 #include "flann_factory.h"
 
+static constexpr int N_TREES = 4;
+static constexpr int N_CHECKS = 40;
 
 phg::FlannMatcher::FlannMatcher()
 {
     // параметры для приближенного поиска
-//    index_params = flannKdTreeIndexParams(TODO);
-//    search_params = flannKsTreeSearchParams(TODO);
+   index_params = flannKdTreeIndexParams(N_TREES);
+   search_params = flannKsTreeSearchParams(N_CHECKS);
 }
 
 void phg::FlannMatcher::train(const cv::Mat &train_desc)
@@ -17,5 +23,14 @@ void phg::FlannMatcher::train(const cv::Mat &train_desc)
 
 void phg::FlannMatcher::knnMatch(const cv::Mat &query_desc, std::vector<std::vector<cv::DMatch>> &matches, int k) const
 {
-    throw std::runtime_error("not implemented yet");
+    cv::Mat idx(query_desc.rows, k, CV_32SC1);
+    cv::Mat dst(query_desc.rows, k, CV_32FC1);
+    flann_index->knnSearch(query_desc, idx, dst, k, *search_params);
+    matches.resize(query_desc.rows);
+    for (int i = 0; i < matches.size(); i++) {
+        matches[i].reserve(k);
+        for (int j = 0; j < k; ++j) {
+            matches[i].push_back(cv::DMatch(i, idx.at<int>(i, j), dst.at<float>(i, j)));
+        }
+    }
 }

src/phg/sfm/homography.cpp

@@ -1,9 +1,13 @@
 #include "homography.h"
 
+#include <cmath>
+#include <functional>
 #include <opencv2/calib3d/calib3d.hpp>
 #include <iostream>
+#include <opencv2/core.hpp>
 
 namespace {
+    constexpr float MIN_DST2 = 1e-6f;
 
     // источник: https://e-maxx.ru/algo/linear_systems_gauss
     // очень важно при выполнении метода гаусса использовать выбор опорного элемента: об этом можно почитать в источнике кода
@@ -84,8 +88,8 @@ namespace {
             double w1 = ws1[i];
 
             // 8 elements of matrix + free term as needed by gauss routine
-//            A.push_back({TODO});
-//            A.push_back({TODO});
+            A.push_back({ w1 * x0, w1 * y0, w1 * w0, 0, 0, 0, -x1 * x0, -x1 * y0, x1 * w0 });
+            A.push_back({ 0, 0, 0, -w1 * x0, -w1 * y0, -w1 * w0, y1 * x0, y1 * y0, -y1 * w0 });
         }
 
         int res = gauss(A, H);
@@ -136,7 +140,7 @@ namespace {
         return *state = x;
     }
 
-    void randomSample(std::vector<int> &dst, int max_id, int sample_size, uint64_t *state)
+    void randomSample(std::vector<int> &dst, int max_id, int sample_size, uint64_t *state, const std::function<bool(int)> &check_new_element)
     {
         dst.clear();
 
@@ -145,7 +149,7 @@ namespace {
         for (int i = 0; i < sample_size; ++i) {
             for (int k = 0; k < max_attempts; ++k) {
                 int v = xorshift64(state) % max_id;
-                if (dst.empty() || std::find(dst.begin(), dst.end(), v) == dst.end()) {
+                if (dst.empty() || std::find(dst.begin(), dst.end(), v) == dst.end() && check_new_element(v)) {
                     dst.push_back(v);
                     break;
                 }
@@ -168,57 +172,67 @@ namespace {
         // * (простое описание для понимания)
         // * [3] http://ikrisoft.blogspot.com/2015/01/ransac-with-contrario-approach.html
 
-//        const int n_matches = points_lhs.size();
-//
-//        // https://en.wikipedia.org/wiki/Random_sample_consensus#Parameters
-//        const int n_trials = TODO;
-//
-//        const int n_samples = TODO;
-//        uint64_t seed = 1;
-//        const double reprojection_error_threshold_px = 2;
-//
-//        int best_support = 0;
-//        cv::Mat best_H;
-//
-//        std::vector<int> sample;
-//        for (int i_trial = 0; i_trial < n_trials; ++i_trial) {
-//            randomSample(sample, n_matches, n_samples, &seed);
-//
-//            cv::Mat H = estimateHomography4Points(points_lhs[sample[0]], points_lhs[sample[1]], points_lhs[sample[2]], points_lhs[sample[3]],
-//                                                  points_rhs[sample[0]], points_rhs[sample[1]], points_rhs[sample[2]], points_rhs[sample[3]]);
-//
-//            int support = 0;
-//            for (int i_point = 0; i_point < n_matches; ++i_point) {
-//                try {
-//                    cv::Point2d proj = phg::transformPoint(points_lhs[i_point], H);
-//                    if (cv::norm(proj - cv::Point2d(points_rhs[i_point])) < reprojection_error_threshold_px) {
-//                        ++support;
-//                    }
-//                } catch (const std::exception &e)
-//                {
-//                    std::cerr << e.what() << std::endl;
-//                }
-//            }
-//
-//            if (support > best_support) {
-//                best_support = support;
-//                best_H = H;
-//
-//                std::cout << "estimateHomographyRANSAC : support: " << best_support << "/" << n_matches << std::endl;
-//
-//                if (best_support == n_matches) {
-//                    break;
-//                }
-//            }
-//        }
-//
-//        std::cout << "estimateHomographyRANSAC : best support: " << best_support << "/" << n_matches << std::endl;
-//
-//        if (best_support == 0) {
-//            throw std::runtime_error("estimateHomographyRANSAC : failed to estimate homography");
-//        }
-//
-//        return best_H;
+       const int n_matches = points_lhs.size();
+
+       // https://en.wikipedia.org/wiki/Random_sample_consensus#Parameters
+       const int n_trials = std::log(1e-3)/std::log(1 - std::pow(0.2, 4));
+
+       const int n_samples = 4;
+       uint64_t seed = 1;
+       const double reprojection_error_threshold_px = 2;
+
+       int best_support = 0;
+       cv::Mat best_H;
+
+       std::vector<int> sample;
+       for (int i_trial = 0; i_trial < n_trials; ++i_trial) {
+           randomSample(sample, n_matches, n_samples, &seed, [&sample, &points_lhs, &points_rhs](int x) {
+                for (int prev : sample) {
+                    auto left_vec = (points_lhs[prev] - points_lhs[x]);
+                    if (std::hypot(left_vec.x, left_vec.y) < MIN_DST2)
+                        return false;
+                    auto right_vec = (points_rhs[prev] - points_rhs[x]);
+                    if (std::hypot(right_vec.x, right_vec.y) < MIN_DST2)
+                        return false;
+                }
+                return true;
+           });
+
+           cv::Mat H = estimateHomography4Points(points_lhs[sample[0]], points_lhs[sample[1]], points_lhs[sample[2]], points_lhs[sample[3]],
+                                                 points_rhs[sample[0]], points_rhs[sample[1]], points_rhs[sample[2]], points_rhs[sample[3]]);
+
+           int support = 0;
+           for (int i_point = 0; i_point < n_matches; ++i_point) {
+               try {
+                   cv::Point2d proj = phg::transformPoint(points_lhs[i_point], H);
+                   if (cv::norm(proj - cv::Point2d(points_rhs[i_point])) < reprojection_error_threshold_px) {
+                       ++support;
+                   }
+               } catch (const std::exception &e)
+               {
+                   std::cerr << e.what() << std::endl;
+               }
+           }
+
+           if (support > best_support) {
+               best_support = support;
+               best_H = H;
+
+               std::cout << "estimateHomographyRANSAC : support: " << best_support << "/" << n_matches << std::endl;
+
+               if (best_support == n_matches) {
+                   break;
+               }
+           }
+       }
+
+       std::cout << "estimateHomographyRANSAC : best support: " << best_support << "/" << n_matches << std::endl;
+
+       if (best_support == 0) {
+           throw std::runtime_error("estimateHomographyRANSAC : failed to estimate homography");
+       }
+
+       return best_H;
     }
 
 }
@@ -238,7 +252,10 @@ cv::Mat phg::findHomographyCV(const std::vector<cv::Point2f> &points_lhs, const
 // таким преобразованием внутри занимается функции cv::perspectiveTransform и cv::warpPerspective
 cv::Point2d phg::transformPoint(const cv::Point2d &pt, const cv::Mat &T)
 {
-    throw std::runtime_error("not implemented yet");
+    double x = T.at<double>(0, 0) * pt.x + T.at<double>(0, 1) * pt.y + T.at<double>(0, 2);
+    double y = T.at<double>(1, 0) * pt.x + T.at<double>(1, 1) * pt.y + T.at<double>(1, 2);
+    double w = T.at<double>(2, 0) * pt.x + T.at<double>(2, 1) * pt.y + T.at<double>(2, 2);
+    return { x / w, y / w };
 }
 
 cv::Point2d phg::transformPointCV(const cv::Point2d &pt, const cv::Mat &T) {