C++ GStreamer SDK as input/outpu

[MrDaniel]

Hello,

Presently, i have a c++ application that does face detection in OpenCV.

Although GStreamer has an OpenCV plugin, i have some custom code that
wouldn't be easy to port.

So...

Input -> Process -> Output

I can use local videos and images, OpenCV also has RTSP but it is buggy.

Each frame is read and passed to the face recognition process.

Using an OpenCV image container, as CvMat, it is possible to use gStreamer.

CvMat frame; // Create a frame
frame->imageData = (char *)GST_BUFFER_DATA(buffer);

This is taken from here.
https://community.freescale.com/thread/319188

I am seeking advice, examples or general pointers for these two questions.

1) How do i make this work for RTMP, RTSP or HLS source?
2) How would i record something frame by frame?

Any advice or guidance 



--
View this message in context: http://gstreamer-devel.966125.n4.nabble.com/C-GStreamer-SDK-as-input-outpu-tp4667665.html
Sent from the GStreamer-devel mailing list archive at Nabble.com.