how to use SIFT in opencv

Question

I am learning C++ and OpenCV these days. Given an image, I want to extract its SIFT features. From http://docs.opencv.org/modules/nonfree/doc/feature_detection.html, we can know that OpenCV 2.4.8 has the SIFT module. See here:

But I do not know how to use it. Currently, to use SIFT, I need to first call the class SIFT to get a SIFT instance. Then, I need to use SIFT::operator()() to do SIFT.

But what is OutputArray , InputArray, KeyPoint? Could anyone give a demo to show how to use SIFT class to do SIFT?

Answer 1

Update for OpenCV 4.2.0 (don’t forget to link opencv_xfeatures2d420.lib, of course)

#include <opencv2/core.hpp>
#include <opencv2/highgui.hpp>
#include <opencv2/xfeatures2d.hpp>

int main(int argc, char** argv)
{
    const cv::Mat input = cv::imread("input.jpg", 0); //Load as grayscale

    cv::Ptr<cv::xfeatures2d::SIFT> siftPtr = cv::xfeatures2d::SIFT::create();
    std::vector<cv::KeyPoint> keypoints;
    siftPtr->detect(input, keypoints);

    // Add results to image and save.
    cv::Mat output;
    cv::drawKeypoints(input, keypoints, output);
    cv::imwrite("sift_result.jpg", output);it.

    return 0;
}

Answer 2

update for OpenCV3

#include <opencv2/core/core.hpp>
#include <opencv2/highgui/highgui.hpp>
#include <opencv2/nonfree/features2d.hpp> //Thanks to Alessandro

int main(int argc, const char* argv[])
{
    const cv::Mat input = cv::imread("input.jpg", 0); //Load as grayscale

    cv::Ptr<cv::SiftFeatureDetector> detector = cv::SiftFeatureDetector::create();
    std::vector<cv::KeyPoint> keypoints;
    detector->detect(input, keypoints);

    // Add results to image and save.
    cv::Mat output;
    cv::drawKeypoints(input, keypoints, output);
    cv::imwrite("sift_result.jpg", output);

    return 0;
}

Answer 3

See the example from Sift implementation with OpenCV 2.2

#include <opencv2/core/core.hpp>
#include <opencv2/highgui/highgui.hpp>
#include <opencv2/nonfree/features2d.hpp> //Thanks to Alessandro

int main(int argc, const char* argv[])
{
    const cv::Mat input = cv::imread("input.jpg", 0); //Load as grayscale

    cv::SiftFeatureDetector detector;
    std::vector<cv::KeyPoint> keypoints;
    detector.detect(input, keypoints);

    // Add results to image and save.
    cv::Mat output;
    cv::drawKeypoints(input, keypoints, output);
    cv::imwrite("sift_result.jpg", output);

    return 0;
}

Tested on OpenCV 2.4.8