How can I only detect objects that are located in the center region of the screen?

Autonomous Machines Jetson & Embedded Systems Jetson Nano
1

Hi, I’m only trying to detect things that are located in the center of the screen. I know this can be done with openCV, but I’m not sure how to do it with detectnet. Any input would be greatly appreciated.

2

You can crop your region of interest and only feed that crop to the engine.

3

Hi, thanks for the response. How would I go about doing that in detectnet.py?

5

Hi @justin60, you can use the cudaCrop() function:

https://github.com/dusty-nv/jetson-inference/blob/master/docs/aux-image.md#cropping

8

Thanks @dusty_nv for the reply. Is that function only for images? Or can it be used for a live video stream as well?

9

I figured it out, thanks for everyone’s help!

10

Glad you got it working! And to answer your last question for others who may read this thread, like most functions in jetson-inference it operates on individual images. However you can just call it for each new image from the video stream as they are captured.

12

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.