Best way to do CV with a Raspberry Pi?

My team recently acquired a Raspberry Pi 4 to use for tracking Power Cells during autonomous. What’s the best way to do this? We were looking at using the FRCVision image for the Pi 4 but were having issues figuring it out. (Pointers to any tutorials in greater detail than the WPILib documentation would be greatly appreciated)
What would you recommend we use?

FRCVision Image was well supported for us last year in prototypes. I hear good things this year about here.

1 Like

Thanks for the reply!
Chameleon-Vision looks great, I am going to download it right now.

How were you able to get FRCVision working? I really don’t understand how to use an OpenCV pipeline with it…
maybe I am just being stupid.

Basically, I just followed the instructions, downloaded the example. Modified it with some GRIP processing and uploaded it.

Worked fairly well out of the box in my testing. I was working against the 2019 image at the beginning of build season last year when our Limelight was on back order.

I don’t have any experience with this year’s image changes to know.

Check into the CV discord. Lots of users and devs on there who can point you in the right direction. Our programmers are having a lot of success with it this year,

This topic was automatically closed 365 days after the last reply. New replies are no longer allowed.