CodeProject.AI Version 2.0

Anyone else seeing this. All of a sudden everything is coming back object not found and times were shooting to 30,000+ms.

This is what my display shows now - I tried stopping the service and restarting the computer and same problem persists of saying Lost Contact - it will make connection and then lose contact again.

View attachment 183729
Do you have Auto start/stop with Blue Iris checked? If so uncheck it.

1705966396168.png
 
Anyone else seeing this. All of a sudden everything is coming back object not found and times were shooting to 30,000+ms.

This is what my display shows now - I tried stopping the service and restarting the computer and same problem persists of saying Lost Contact - it will make connection and then lose contact again.

View attachment 183729
Getting the same issue. I'd have probably 900-1500 requests then lose connection and it'd restart.
 
Anyone else seeing this. All of a sudden everything is coming back object not found and times were shooting to 30,000+ms.

This is what my display shows now - I tried stopping the service and restarting the computer and same problem persists of saying Lost Contact - it will make connection and then lose contact again.

View attachment 183729
I get this on every restart but it settles after a couple of minutes. Very high CPU and memory usage for the CPAI process while it happens (100% CPU and 2Gb+ memory usage for the process). Runs smoothly with GPU DirectML thereafter using 250Mb memory and less than 5% CPU. The 30 second duration is likely a timeout? I'm running 2.4.7 on an i7-11700:

1705979142123.png
 
Not sure if anyone noticed this however all versions starting from 2.5.0 cause ai error 500 log entries in BI the detection still works though and it seems like a timeout issue.

I’m only using the coral module and reverting to 2.4.7 fixes it I’m using the single tpu code
 
  • Like
Reactions: David L
Hey.
Anyone noticed Nivdia gpu’s starting/stopping when using yolov5.NET? Seems to prefer yolo.6.2. If I leave yolo.6.2 installed and stop it. I can get yolo.net to run and stay running. If I uninstall yolo.6.2 and only have yolo.net it acts up. Prefer Yolo.net for the speed. Tried with 2.5.0 and New 2.5.1.
 
  • Like
Reactions: David L
Since CodeProject.AI is out of beta I started a new thread for all topics CodeProject.AI version 2.5 and greater.

 
Last edited:
Hi

I have BI setup with approximately 15cams running (Hikvisions) some of these are clone cams for AI alerts. I am detecting 'person'only using ipcam-combined.

Is this the best model to use with YOLOv5.Net 1.9.2? or is there a faster way to process?

I typically set to detect 5 images at 500ms

I notice then when triggered and detecting CPU spikes to 100% for a few seconds and then drops down to around 15%

Is the spiking normal or is there something I can do to assist this. I have a Lenovo AIO pc with 16gb Ram

CPU - i5-10400T @ 2.00GHz
6cores 12 processors
Intel UHD 630 GPU


Status logs in CPAI report around 250ms -1500ms at time or higher

Also get Random shutdowns

Use Main Stream is unticked

Any advice/help is appreciated

thanks
 
  • Like
Reactions: David L
@ch3st3rk Given the details you shared, since you unticked main stream, BI5 should send substream image to CP.AI. Is substream configrued for that camera?

Also, please share screenshots for:
#1: AI tab for that camera
#2: The AI details for a triggered event.

Thanks, I have substreams off on BI so I can view in better quality. Hope the attached help.

Vapour from my boiler causes a load of CPU jump!
 

Attachments

  • BlueIris_9aagyMmF9d.png
    BlueIris_9aagyMmF9d.png
    44.2 KB · Views: 26
  • BlueIris_30YCkqCfSU.png
    BlueIris_30YCkqCfSU.png
    40.3 KB · Views: 27
  • BlueIris_CBf6qbAJnd.png
    BlueIris_CBf6qbAJnd.png
    31.3 KB · Views: 23
  • Like
Reactions: David L
@ch3st3rk If you don't have substream configured for camera, then BI5 has no choice but to use main stream image for CP.AI. Main stream image will take more CPU to do AI detect than substream image. Add substream to camera temporarily to see if it improves AI response time.

BTW, please send #2 requested screenshot, the AI analysis details for a triggered event. It looks like this:
ai analysis.png
To get this screen, when in BI5 console, on your left, hold down Ctrl key while double clicking on a triggered event thumbnail

Note the AI response time of <100ms in mine, and the image resolution in the upper right corner.
 
  • Like
Reactions: David L
@ch3st3rk If you don't have substream configured for camera, then BI5 has no choice but to use main stream image for CP.AI. Main stream image will take more CPU to do AI detect than substream image. Add substream to camera temporarily to see if it improves AI response time.

BTW, please send #2 requested screenshot, the AI analysis details for a triggered event. It looks like this:
View attachment 184884
To get this screen, when in BI5 console, on your left, hold down Ctrl key while double clicking on a triggered event thumbnail

Note the AI response time of <100ms in mine, and the image resolution in the upper right corner.
Thanks I will have to take a look a little later, will using sub stream prevent the CPU spikes or is that quite normal when detection triggered to analyse AI object?
thanks
 
  • Like
Reactions: David L
@ch3st3rk It's best practice to configure main+sub streams for every camera to reduce CPU load, especially if you have cameras with high resolution like 4k. You're not missing out, you still have access to the main streams for viewing.

In general, AI models are trained on lower resolution images so if your images are high-res like 4k, they will be downsized (takes time and CPU load) as part of AI detection.
 
Last edited:
@ch3st3rk It's best practice to configure main+sub streams for every camera to reduce CPU load, especially if you have cameras with high resolution like 4k. You're not missing out, you still have access to the main streams for viewing.

In general, AI models are trained on lower resolution images so if your images are high-res like 4k, they will be downsized (takes time and CPU load) as part of AI detection.
Ok thanks I will setup some sub streams and see how that goes and report back, thanks for your help!
 
  • Like
Reactions: actran and David L