Recent content by devastator

Still don't see it. Not sure what your problem is. Couldn't bother less to be honest.

Relax man, I see no false assertions here ...

This is particularly hard, because e.g. for objects further away something is detected in frame 1, but not anymore in frame 10 (noise, lighting conditions, ...). If you want something like this, you would defintely need object tracking. I saw this with my Google Coral implementation too (and am...

That is what I did. Thanks for the excellent tutorial. However, I've bought 10+ cams from a reseller in the region, just to avoid this mess. I knew the problems with chinese hacked versions, and i wanted to avoid that. Otherwise I could have just sold em directly on ali or the like

SADP didn't show anything. After flashing the CN version, it showed the update complete message. Guess I will contact my reseller why he sold me a chinese version.

I bought the my DS-2CD2732F-IS from an authorized reseller in the Netherlands. I tried to upgrade yesterday (downloaded from the American website), but after the upgrade the camera doesn't come online anymore. I downloaded TFTP and the european SW version (are they different), and managed to...

No, because that does not process all frames. My script is working on the RTSP stream.

I've been messing with Coral USB Accelerator to process every single frame. I've bumped my confidence to > 40%. Sure, it will miss persons in some frames, but in the end it will always have identified it in one of the others... Will share here once it's ready for prime time.

Hi, I'm trying to parse the video camera feeds to a Google Coral USB Accelerator, in order to do object detection. The main profile of my cams are H264, the substreams are MJPEG (but have a much lower resolution). I know BlueIris also has MJPEG streams, but they do not have the same resolution...

It is not integrated yet. I just have it running next to blueiris for now.

Also, it is not that big of a deal. I can process nearly every single frame now, so i can compensate for accuracy if needed.

TF lite. Have had no issues yet. I am doing a fait bit of post processing though (e.g. non maximum suppresion, object has to be found in 3 consecutive frames, ...)

Got my Google Coral USB Accelerator yesterday. I currently have it parsing the rtsp streams. Getting detection times of <30 ms.

In the AI tool you can mask part of the image. I think this would solve your problem ...

HomeAssistant is not required, but it is more fun when you have HA coupled with your cameras :-)