For anyone curious, this line of comment shows how it's done:
> startCameraAndStream opens the webcam with GoCV, sends raw frames to FFmpeg (via stdin), reads IVF from FFmpeg (via stdout), and writes them into the WebRTC video track.
Neat demo — cool to see GoCV + Pion working together over WebRTC. Curious if you’ve tried running object detection or overlays before sending the stream?
I did! I tried integrating with a couple different of the examples on here [0]
The good detection ones required users to download something first, so I was worried the friction would make people less likely to use them. The one I almost went with was hand-gestures, but it wasn't as reliable.
I haven't used opencv that much. I see people talk about in Pion discord (and examples like this) and gets me into. I think someone more proficient could build a more compelling demo pretty quickly :)
I used to do some basic CV stuff back when I did the First Robotics Competition (FRC) in high school, so it's cool to see it being done over WebRTC!
Hey good to see you, we gotta get your PR merged and on here also :)
It's really cool how you can have a heavy server doing all this cool computer vision stuff, but then a simple client that just does WebRTC.
If you adjusted the example you can quickly be sending OBS or ffmpeg in. Then that gets really cool/powerful
Yeah the PR is pretty much "done" at this point (there's more stuff I could probably do but perfect is the enemy of good)
For anyone curious, this line of comment shows how it's done:
> startCameraAndStream opens the webcam with GoCV, sends raw frames to FFmpeg (via stdin), reads IVF from FFmpeg (via stdout), and writes them into the WebRTC video track.
If people want to build it another way you could do any of these, examples all on the repo!
* Use FFMPEG lib directly * GStreamer * Use libvpx, examples is in mediadevices repo
Neat demo — cool to see GoCV + Pion working together over WebRTC. Curious if you’ve tried running object detection or overlays before sending the stream?
I did! I tried integrating with a couple different of the examples on here [0]
The good detection ones required users to download something first, so I was worried the friction would make people less likely to use them. The one I almost went with was hand-gestures, but it wasn't as reliable.
I haven't used opencv that much. I see people talk about in Pion discord (and examples like this) and gets me into. I think someone more proficient could build a more compelling demo pretty quickly :)
[0] https://gocv.io/writing-code/more-examples/
Huh, wonder if go2rtc could use this to make HKSV cameras work.
Consepcion jimenez vinagre