How to convert gstCamera stream to opencv frame?

Autonomous Machines Jetson & Embedded Systems Jetson Nano
1

Thanks in advanced. In jetson-inference, I want to use opencv in detectnet-camera.cpp, the first thing is convert gstCamera stream to opencv frame, as code below:

float confidence = 0.0f;

	while (!signal_recieved)
	{
		// capture RGBA image
		float *imgRGBA = NULL;

		if (!camera->CaptureRGBA(&<b>imgRGBA</b>, 1000))
			printf("detectnet-camera:  failed to capture RGBA image from camera\n");

		// detect objects in the frame
		detectNet::Detection *detections = NULL;

Obviously, should do something about imgRGBA, would any expert tell me how to handle it?

2

Hi,
imgRGBA is in float which may not work in OpenCV. You may need imgRGBA to be uint8_t.
This is not right. Please refer to #3.

3

Hi,
Please check the following post:
https://devtalk.nvidia.com/default/topic/1065260/jetson-nano/jetson-inference-with-opencv/post/5394370/#5394370