opencv4 mask_rcnn模型调(c++)

昨天有人问我关于调用mask_rcnn模型的问题，忽然想到最近三个月都没用opencv调用训练好的mask_rcnn模型了，今晚做个尝试，所以重新编译了 opencv4，跑个案例试试

#include <fstream>
#include <sstream>
#include <iostream>
#include <string.h>

#include <opencv2/dnn.hpp>
#include <opencv2/imgproc.hpp>
#include <opencv2/highgui.hpp>

using namespace cv;
using namespace dnn;
using namespace std;

RNG rng1;

// Initialize the parameters
float confThreshold = 0.5; // Confidence threshold
float maskThreshold = 0.3; // Mask threshold

//vector<string> classes;
//vector<Scalar> colors;

// Draw the predicted bounding box
void drawBox(Mat& frame, int classId, float conf, Rect box, Mat& objectMask);

// Postprocess the neural network's output for each frame
void postprocess(Mat& frame, const vector<Mat>& outs);

int main()
{
    // Give the configuration and weight files for the model
    //String textGraph = "./mask_rcnn_inception_v2_coco_2018_01_28/mask_rcnn_inception_v2_coco_2018_01_28.pbtxt";
    //String modelWeights = "./mask_rcnn_inception_v2_coco_2018_01_28/frozen_inference_graph.pb";

    String modelWeights = "E:\\Opencv\\model_1\\mask_rcnn_inception_v2_coco_2018_01_28\\frozen_inference_graph.pb";
    String textGraph = "E:\\Opencv\\model_1\\mask_rcnn_inception_v2_coco_2018_01_28\\mask_rcnn_inception_v2_coco_2018_01_28.pbtxt";
    // Load the network
    Net net = readNetFromTensorflow(modelWeights, textGraph);
    net.setPreferableBackend(DNN_BACKEND_OPENCV);
    net.setPreferableTarget(DNN_TARGET_CPU);

    // Open a video file or an image file or a camera stream.
    string str, outputFile;
    VideoCapture cap();//根据摄像头端口id不同，修改下即可
                        //VideoWriter video;
    Mat frame, blob;

    // Create a window
    static const string kWinName = "Deep learning object detection in OpenCV";
    namedWindow(kWinName, WINDOW_NORMAL);

    // Process frames.
    if (>)
    {
        // get frame from the video
        //cap >> frame;
        frame = cv::imread("D:\\image\\5.png");

        // Stop the program if reached end of video
        if (frame.empty())
        {
            cout << "Done processing !!!" << endl;
            cout << "Output file is stored as " << outputFile << endl;
        }
        // Create a 4D blob from a frame.
        blobFromImage(frame, blob, 1.0, Size(frame.cols, frame.rows), Scalar(), true, false);
        //blobFromImage(frame, blob);

        //Sets the input to the network
        net.setInput(blob);

        // Runs the forward pass to get output from the output layers
        std::vector<String> outNames();
        outNames[] = "detection_out_final";
        outNames[] = "detection_masks";
        vector<Mat> outs;
        net.forward(outs, outNames);

        // Extract the bounding box and mask for each of the detected objects
        postprocess(frame, outs);

        // Put efficiency information. The function getPerfProfile returns the overall time for inference(t) and the timings for each of the layers(in layersTimes)
        vector<double> layersTimes;
        double freq = getTickFrequency() / ;
        double t = net.getPerfProfile(layersTimes) / freq;
        string label = format("Mask-RCNN on 2.5 GHz Intel Core i7 CPU, Inference time for a frame : %0.0f ms", t);
        putText(frame, label, Point(, ), FONT_HERSHEY_SIMPLEX, 0.5, Scalar(, , ));

        // Write the frame with the detection boxes
        Mat detectedFrame;
        frame.convertTo(detectedFrame, CV_8U);

        imshow(kWinName, frame);

    }
    //cap.release();
    waitKey();
    return ;
}

// For each frame, extract the bounding box and mask for each detected object
void postprocess(Mat& frame, const vector<Mat>& outs)
{
    Mat outDetections = outs[];
    Mat outMasks = outs[];

    // Output size of masks is NxCxHxW where
    // N - number of detected boxes
    // C - number of classes (excluding background)
    // HxW - segmentation shape
    const int numDetections = outDetections.size[];
    const int numClasses = outMasks.size[];

    outDetections = outDetections.reshape(, outDetections.total() / );
    for (int i = ; i < numDetections; ++i)
    {
        float score = outDetections.at<float>(i, );
        if (score > confThreshold)
        {
            // Extract the bounding box
            int classId = static_cast<int>(outDetections.at<float>(i, ));
            int left = static_cast<int>(frame.cols * outDetections.at<float>(i, ));
            int top = static_cast<int>(frame.rows * outDetections.at<float>(i, ));
            int right = static_cast<int>(frame.cols * outDetections.at<float>(i, ));
            int bottom = static_cast<int>(frame.rows * outDetections.at<float>(i, ));

            left = max(, min(left, frame.cols - ));
            top = max(, min(top, frame.rows - ));
            right = max(, min(right, frame.cols - ));
            bottom = max(, min(bottom, frame.rows - ));
            Rect box = Rect(left, top, right - left + , bottom - top + );

            // Extract the mask for the object
            Mat objectMask(outMasks.size[], outMasks.size[], CV_32F, outMasks.ptr<float>(i, classId));

            // Draw bounding box, colorize and show the mask on the image
            drawBox(frame, classId, score, box, objectMask);

        }
    }
}

// Draw the predicted bounding box, colorize and show the mask on the image
void drawBox(Mat& frame, int classId, float conf, Rect box, Mat& objectMask)
{
    //Draw a rectangle displaying the bounding box
    rectangle(frame, Point(box.x, box.y), Point(box.x + box.width, box.y + box.height), Scalar(, , ), );

    //Get the label for the class name and its confidence
    /*string label = format("%.2f", conf);
    if (!classes.empty())
    {
        CV_Assert(classId < (int)classes.size());
        label = classes[classId] + ":" + label;
    }*/

    //Display the label at the top of the bounding box
    /*
    int baseLine;
    Size labelSize = getTextSize(label, FONT_HERSHEY_SIMPLEX, 0.5, 1, &baseLine);
    box.y = max(box.y, labelSize.height);
    rectangle(frame, Point(box.x, box.y - round(1.5*labelSize.height)), Point(box.x + round(1.5*labelSize.width), box.y + baseLine), Scalar(255, 255, 255), FILLED);
    putText(frame, label, Point(box.x, box.y), FONT_HERSHEY_SIMPLEX, 0.75, Scalar(0, 0, 0), 1);
*/
    //Scalar color = colors[classId%colors.size()];
    Scalar color = Scalar(rng1.uniform(, ), rng1.uniform(, ), rng1.uniform(, ));

    // Resize the mask, threshold, color and apply it on the image
    resize(objectMask, objectMask, Size(box.width, box.height));
    Mat mask = (objectMask > maskThreshold);
    Mat coloredRoi = (0.3 * color + 0.7 * frame(box));
    coloredRoi.convertTo(coloredRoi, CV_8UC3);

    // Draw the contours on the image
    vector<Mat> contours;
    Mat hierarchy;
    mask.convertTo(mask, CV_8U);
    findContours(mask, contours, hierarchy, RETR_CCOMP, CHAIN_APPROX_SIMPLE);
    drawContours(coloredRoi, contours, -, color, , LINE_8, hierarchy, );
    coloredRoi.copyTo(frame(box), mask);

}

检测速度和python比起来偏慢

运行日志：

[ INFO:0] global E:\Opencv\opencv-4.1.1\modules\videoio\src\videoio_registry.cpp (187) cv::`anonymous-namespace'::VideoBackendRegistry::VideoBackendRegistry VIDEOIO: Enabled backends(7, sorted by priority): FFMPEG(1000); GSTREAMER(990); INTEL_MFX(980); MSMF(970); DSHOW(960); CV_IMAGES(950); CV_MJPEG(940)
[ INFO:0] global E:\Opencv\opencv-4.1.1\modules\videoio\src\backend_plugin.cpp (340) cv::impl::getPluginCandidates Found 2 plugin(s) for GSTREAMER
[ INFO:0] global E:\Opencv\opencv-4.1.1\modules\videoio\src\backend_plugin.cpp (172) cv::impl::DynamicLib::libraryLoad load E:\Opencv\opencv_4_1_1_install\bin\opencv_videoio_gstreamer411_64.dll => FAILED
[ INFO:0] global E:\Opencv\opencv-4.1.1\modules\videoio\src\backend_plugin.cpp (172) cv::impl::DynamicLib::libraryLoad load opencv_videoio_gstreamer411_64.dll => FAILED
[ INFO:0] global E:\Opencv\opencv-4.1.1\modules\core\src\ocl.cpp (888) cv::ocl::haveOpenCL Initialize OpenCL runtime...