Correct way copying depth data or color data to opencvMat

Question

Hello,
I want to collect synchronized IR data, depth data and color data from the kinect v2. Following the pattern from the sample, CoordinateMapper, I am stuck on the section below:
if (imageFormat == ColorImageFormat_Bgra)
hr = pColorFrame-AccessRawUnderlyingBuffer(&nColorBufferSize, reinterpret_cast<BYTE**>(&pColorBuffer));
else if (m_pColorRGBX)
pColorBuffer = m_pColorRGBX;
nColorBufferSize = cColorWidth * cColorHeight * sizeof(RGBQUAD);
hr = pColorFrame<CopyConvertedFrameDataToArray(nColorBufferSize, reinterpret_cast<BYTE*>(pColorBuffer), ColorImageFormat_Bgra);
In particular,
1) why is there a copy from the underlying buffer and
2) what is m_pColorRGBX (initialization for the color image buffer perharps?)
3) how would I copy the resulting color image to an opencv cv::Mat array in a memory safe fashion?
Any help is greatly appreciated.

Answer 1

Hi Carmine Sirignano,
thank you for the information. 
So, if I understand correctly, the code above says,
if the imageFormat of the incoming frame is already in BGRA, THEN->,
Copy the underlying buffer to the colorBuffer
BUT if the m_pColorRGBX is initialized, THEN ->
Initialize the color buffer, and copy into the buffer
I understand the first condition, but I am unclear as to why we have the second condition for m_pColorRGBX. Can you explain why this is?
Finally, I would like to copy the buffer data into a cv::Mat. Can you give any guidelines on how to do this in a safe fashion?