Fetching Realtime buffer data

[SutirthaChakraborty]

Hi,

Thank you for this lovely SDK. I am trying to explore this library.
My intension is to calculate the movement of individuals using pose estimation.

How can I implement PoseCamera on video/webcam data in realtime? And get the estimation value in a small buffer data (maybe of 4 secs ) to further process it, to find the movement in each individual. I am trying to use google colab for it. Any example notebook?

Regards,
Suti

[usmankaiz]

The SDK can be work for a real-time scenario like web cameras and it can also be used with offline videos. What exactly do you mean by a small buffer of 4 sec?

You can collect the joints data from the SDK and store data according to the frame number. it's up to you how you want to store the joints data in a file either JSON or XML.

Hope this helps.

[SutirthaChakraborty]

Thank you for the reply. I understood now.

[SutirthaChakraborty]

I have initialized
WIDTH = 870 HEIGHT = 576 cap = cv2.VideoCapture(0)
When I am running this code. It's too slow to be real-time. Any way to make it fast?

	while(True):

		ret, frame = cap.read()

		gray= cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
		gray= cv2.resize(frame,(870, 576))

		poses= posecamera.estimate(frame )
		for pose in poses:
			pose.draw(gray)

		cv2.imshow('frame',gray)
		if cv2.waitKey(1) & 0xFF == ord('q'):
			break

	cap.release()

[usmankaiz]

Can you share your system specs?

[SutirthaChakraborty]

Hi,
OS: Windows 10
RAM : 16 GB
Memory: 512 GB
Processor: AMD Ryzen 7 3700U with Radeon Vega Mobile Gfx, 2300 Mhz, 4 Core(s), 8 Logical Processor(s)
Integrated GPU

[usmankaiz]

These are some requirements that you need to run the following repository
Python >= 3.6
CUDA 10.1 and Cudnn