Recent content by harleyl7

Good ideas. Thanks! I'll get on the side of my house and change the angle. I threw that one up recently and haven't dialed it in. Good point about the IR light. I was thinking my LPR IR lights was already pushing enough

Yeah true haha the LPR still unfortunately wouldn't have got anything because the only plate we have is on the rear. But I'll be adding a second one. At least this is an excuse to add more to my system haha.

Had a police officer knock on my door this morning and tell me that my neighbors' car was broken into. Seems the car was unlocked, and door checkers are known to come through once the weather starts to warm up. All the clips were combined and some frame rates didn't go well together. Playback...

Ah gotcha. Guess I need to throw a 1060 in my server. But just hold off on LPR for a little. The GTX 1080 just eats too much power

@MikeLud1 do you use the Rockchip for all detection or just LPR? My cpu can't really handle the load of CPAI and I don't want to waste the watts by running my GTX 1080. So will the Rockchip work for LPR and main detection? ~9 cameras, but not all are running AI

For people using CPAI and around an i5-6500 era cpu, what kind of system usage are you seeing?

I've thought about doing something like this, but I'm not ready to move it off my rack currently. Ive thought about buying rack locks since I doubt they're gonna get the full 24u rack out with all the equipment.

It won't work until it's showing under coral driver settings on unRAID. Go to tools > system settings and make sure it's not binded to vfio. It shouldn't have a check next to it. If you were to pass through to a VM then yes we would use vfio, but for this case we want it not to be.

Try this out first https://github.com/HarleyL7/unRAID-CoralTPU-Guide You need to download the coral drivers and pass through the coral to the container.

@MikeLud1 I'm unfortunately getting the same "no predictions returned" issue. .NET with GPU and most recent ALPR module

That's how he's training all these models with that 4090 lol

Nope, I was using an m.2 coral. I think once they implement custom models on CP.AI we will see better times.

My bad I meant to say GTX 1060 3GB. 3GB was enough for most AI detection, but not for ALPR.

Love the GPU times though. I tried the Google coral and it's not there yet, but hopefully it can get down to those times because the power consumption vs gpu is a lot better. Even though GPU doesn't use a ton more power. Currently though I have to use a 1080 with 3 fans so it's a little more...

The CP.AI installer is broken right now even when updating modules. Give it a couple of hours and they will release the new 2.2 one that will fix this . Or if you're up for it there's more complicated ways to get it running now, but honestly not worth it. Files should be getting pushed soon...