Finalized the sample

modules/dnn/samples/ssd_object_detection.cpp

@@ -9,54 +9,42 @@ using namespace cv::dnn;
 #include <cstdlib>
 using namespace std;
 
+const size_t width = 300;
+const size_t height = 300;
 
-//static void colorizeSegmentation(dnn::Blob& score,
-//                                 const vector<cv::Vec3b>& colors,
-//                                 cv::Mat& segm)
-//{
-//    const int rows = score.rows();
-//    const int cols = score.cols();
-//    const int chns = score.channels();
-
-//    cv::Mat maxCl(rows, cols, CV_8UC1);
-//    cv::Mat maxVal(rows, cols, CV_32FC1);
-//    for (int ch = 0; ch < chns; ch++)
-//    {
-//        for (int row = 0; row < rows; row++)
-//        {
-//            const float* ptrScore = score.ptrf(0, ch, row);
-//            uchar* ptrMaxCl = maxCl.ptr<uchar>(row);
-//            float* ptrMaxVal = maxVal.ptr<float>(row);
-//            for (int col = 0; col < cols; col++)
-//            {
-//                if (ptrScore[col] > ptrMaxVal[col])
-//                {
-//                    ptrMaxVal[col] = ptrScore[col];
-//                    ptrMaxCl[col] = ch;
-//                }
-//            }
-//        }
-//    }
-
-//    segm.create(rows, cols, CV_8UC3);
-//    for (int row = 0; row < rows; row++)
-//    {
-//        const uchar* ptrMaxCl = maxCl.ptr<uchar>(row);
-//        cv::Vec3b* ptrSegm = segm.ptr<cv::Vec3b>(row);
-//        for (int col = 0; col < cols; col++)
-//        {
-//            ptrSegm[col] = colors[ptrMaxCl[col]];
-//        }
-//    }
-//}
-
-const char* about = "This sample uses Single-Shot Detector to detect objects "
-                    "from camera\n"; // TODO: link
+Mat getMean(const size_t& height, const size_t& width)
+{
+    Mat mean;
+
+    const int meanValues[3] = {104, 117, 123};
+    vector<Mat> meanChannels;
+    for(size_t i = 0; i < 3; i++)
+    {
+        Mat channel(height, width, CV_32F, Scalar(meanValues[i]));
+        meanChannels.push_back(channel);
+    }
+    cv::merge(meanChannels, mean);
+    return mean;
+}
+
+void preprocess(Mat& frame)
+{
+    frame.convertTo(frame, CV_32FC3);
+    resize(frame, frame, Size(width, height)); //SSD accepts 300x300 RGB-images
+
+    Mat mean = getMean(width, height);
+    cv::subtract(frame, mean, frame);
+}
+
+const char* about = "This sample uses Single-Shot Detector "
+                    "(https://arxiv.org/abs/1512.02325)"
+                    "to detect objects on image\n"; // TODO: link
 
 const char* params
-    = "{ help    | help                | false | print usage         }"
-      "{ proto   | model prototxt file |       | model configuration }"
-      "{ model   | caffemodel file     |       | model weights       }";
+    = "{ help   | help                | false | print usage         }"
+      "{ proto  | model prototxt file |       | model configuration }"
+      "{ model  | caffemodel file     |       | model weights       }"
+      "{ image  | image file          |       | image for detection }";
 
 int main(int argc, char** argv)
 {
@@ -102,53 +90,50 @@ int main(int argc, char** argv)
     importer.release();          //We don't need importer anymore
     //! [Initialize network]
 
-    VideoCapture camera;
-    if (!camera.open(0))
-    {
-        cout << "Unable to open camera stream" << endl;
-        return 0;
-    }
+    cv::Mat frame = cv::imread(parser.get<string>("image"), -1);
 
-    size_t i = 0;
-    for (;; )
+    //! [Prepare blob]
+    preprocess(frame);
+
+    dnn::Blob inputBlob = dnn::Blob(frame);           //Convert Mat to dnn::Blob image
+    //! [Prepare blob]
+
+    //! [Set input blob]
+    net.setBlob(".data", inputBlob);                //set the network input
+    //! [Set input blob]
+
+    //! [Make forward pass]
+    net.forward();                                  //compute output
+    //! [Make forward pass]
+
+    //! [Gather output]
+    dnn::Blob detection = net.getBlob("detection_out");
+    Mat detectionMat(detection.rows(), detection.cols(), CV_32F, detection.ptrf());
+
+    for(size_t i = 0; i < detectionMat.rows; i++)
     {
-        Mat frame;
-        camera >> frame;
-        if (frame.empty())
-            break;
-
-        //! [Prepare blob]
-        resize(frame, frame, Size(300, 300));       //SSD accepts 300x300 RGB-images
-        dnn::Blob inputBlob = dnn::Blob(frame);       //Convert Mat to dnn::Blob image
-        //! [Prepare blob]
-
-        std::ostringstream stream;
-        stream << "folder/" << i << ".jpg";
-        imwrite(stream.str(), frame);
-
-        //! [Set input blob]
-        net.setBlob(".data", inputBlob);            //set the network input
-        //! [Set input blob]
-
-        //! [Make forward pass]
-        net.forward();                              //compute output
-        //! [Make forward pass]
-
-//        //! [Gather output]
-//        dnn::Blob detection = net.getBlob("detection_out");
-
-//        //    cv::Mat colorize;
-//        //    colorizeSegmentation(score, colors, colorize);
-//        //    cv::Mat show;
-//        //    cv::addWeighted(img, 0.4, colorize, 0.6, 0.0, show);
-//        //    cv::imshow("show", show);
-//        //    cv::waitKey(0);
-//        //    return 0;
-
-//        imshow("frame", frame);
-//        if (waitKey(1) == 27)
-//            break;  // stop capturing by pressing ESC
+        std::cout << "Class: " << detectionMat.at<float>(i, 1) << std::endl;
+        std::cout << "Confidence: " << detectionMat.at<float>(i, 2) << std::endl;
+
+        std::cout << " " << detectionMat.at<float>(i, 3) * width;
+        std::cout << " " << detectionMat.at<float>(i, 4) * height;
+        std::cout << " " << detectionMat.at<float>(i, 5) * width;
+        std::cout << " " << detectionMat.at<float>(i, 6) * height;
+
+        float xLeftBottom = detectionMat.at<float>(i, 3) * width;
+        float yLeftBottom = detectionMat.at<float>(i, 4) * height;
+        float xRightTop = detectionMat.at<float>(i, 5) * width;
+        float yRightTop = detectionMat.at<float>(i, 6) * height;
+
+        Rect object(xLeftBottom, yLeftBottom,
+                    xRightTop - xLeftBottom,
+                    yRightTop - yLeftBottom);
+
+        rectangle(frame, object, Scalar(0, 255, 0));
     }
 
-    camera.release();
+    imshow("detections", frame);
+    waitKey();
+
+    return 0;
 } // main