Conversion between dlib::array2d<dlib::rgb_pixel> and qt::QImage.

[Facundo Jose Cancino]

Hello. I am creating a program with QT and using dlib for object detection. In some point I need to show an image via QT generated GUI, so I want to convert a dlib::array2d<dlib::rgb_pixel> to qt::QImage. Any suggestion on how to do that? I am stuck. An alternative could be conversion to Opencv image format, cv::Mat, and then converting cv::Mat to QImage, but I want to avoid that unnecesary step. Anyway, help converting dlib's format to cv::Mat will be apreciated too, I found in this forum information regarding only the inverse convertion.
Thanks in advance.</dlib::rgb_pixel>

[Davis]

You can use dlib::toMat() to convert to opencv. I've never used Qt's image
object so I don't know what you would use there.

[Evgeniy]

Hi, you can use QImage contructor that creates image from memory buffer:
http://doc.qt.io/qt-5/qimage.html#QImage-6

To get pointer from array2d you can look into dlib/array2d/array2d_generic_image.h and use dlib::image_data function

QImage's constructor does not copy memory buffer, so you should make array2d to exists in memory until QImage's destructor called - QImage has cleanup function for it and you can find samples on stackoverflow

[stan-sack]

I don't know if you ever solved this but its probably because of the order of your #includes. Move all the opencv and dlib stuff before foundation.h

[mehrdad shoebyri]

you can use https://github.com/mshobeyri/QFaceRecognition repo that is a qt/qml wrapper on dlib.