# Revision history [back]

You will need to start for sure with the following option.

VideoCapture cap(0 + CV_CAP_INTELPERC)


This will force the capture interface to use the intel interface. Then it is kind of a wasteland out there. Multiple cap interfaces which are poorly documented. I would take a closer look at the sourcecode found here. I notice that there is a get and set method and that there are parameters CV_CAP_INTELPERC_IMAGE and CV_CAP_INTELPERC_DEPTHMAP so it will be possible to retrieve them there I think.

You will need to start for sure with the following option.

VideoCapture cap(0 + CV_CAP_INTELPERC)


This will force the capture interface to use the intel interface. Then it is kind of a wasteland out there. Multiple cap interfaces which are poorly documented. I would take a closer look at the sourcecode found here. I notice that there is a get and set method and that there are parameters CV_CAP_INTELPERC_IMAGE and CV_CAP_INTELPERC_DEPTHMAP so it will be possible to retrieve them there I think.

Especially look at this code line! It seems that you can specify which frame you want to retrieve using the interface!