[tool] [tutorial] Free AI Person Detection for Blue Iris

The other process error is because BI hasn’t quite finished writing the file so has it locked open, if you look it retries and then succeeds.

What is strange are try 2 separate requests to DQ, 1 using an IP address, the other using local host but both reference different ports.

Do you have 2 instances of DQ running?
 
The other process error is because BI hasn’t quite finished writing the file so has it locked open, if you look it retries and then succeeds.

What is strange are try 2 separate requests to DQ, 1 using an IP address, the other using local host but both reference different ports.

Do you have 2 instances of DQ running?
Don't think

I gave up and went back to windows deepstack.. would like to use docker to a save cpu tho
 
Last edited:
Hmmm...just installed a camera in the garage, but Deepstack will not seem to figure out "person". Is it because the cars take up too much of the image? Strange thing is that I'm the only "thing" moving in these pics. Noticed it last night as well when my wife came home, isn't triggering the camera at all (I have car turned off and only person on).

Seems to be the only camera with this issue.

1594910837353.png
1594910932105.png
 
I also found this, it sometimes just fails to spot a person on my driveway too. Having looked at my snapshots I think it might be to do with the overall contrast, my son has a black car and if that's on the driveway it misses people but if my wifes car is parked there instead it seems to detect better, the only real difference is my wifes car is light grey.

In order to resolve these misses I've also configured a motion zone for the cams but confined the zone to a specific small area, so this acts like a backup.
 
Don't think

I gave up and went back to windows deepstack.. would like to use docker to a save cpu tho

I PM'ed you but this is the command line I used:

docker run —name="deepstack" -e TZ=Europe/London -e VISION-DETECTION=True -v localstorage:/datastore -p 8083:5000 --restart always deepquestai/deepstack

Change the port to whichever you use and TZ to your TimeZone.
 
Last edited:
Thanks, so just so I get it:

1. Do we need to insert credentials in the login tab for each (i.e., select "this account") or just leave it as local system account?
2. For python.exe, do we add exactly this into the arguments cell below: ../intelligence.py -MODE=Medium -VFACE=False -VSCENE=False -VDETECTION=True
3. For server.exe, is this what we add into the arguments cell: c../intelligence.py -MODE=Medium -VFACE=False -VSCENE=False -VDETECTION=True

Not sure if the ".." in the python argument and the initial quotes in the server.exe argument were there on purpose.

4. Anything else required in any of the other tabs?

View attachment 66455

View attachment 66456
1. Yes, i run the services as my normal user
2. I have this in arguments for python: C:\DeepStack\intelligence.py -MODE=Medium -VFACE=False -VSCENE=False -VDETECTION=True
2b. I have this in the IO tab for output and error: C:\DeepStack\Logs\pyIntelligence.log
2c. On File rotation I have enabled: Replace existing files
3. I have this for arguments: -VISION-FACE=False -VISION-SCENE=False -VISION-DETECTION=True -ADMIN-KEY= -API-KEY= -PORT=88
3b. I have this in the IO tab for output and error: C:\DeepStack\Logs\DeepStack.log
3c. Same as 2c
 
1. Yes, i run the services as my normal user
2. I have this in arguments for python: C:\DeepStack\intelligence.py -MODE=Medium -VFACE=False -VSCENE=False -VDETECTION=True
2b. I have this in the IO tab for output and error: C:\DeepStack\Logs\pyIntelligence.log
2c. On File rotation I have enabled: Replace existing files
3. I have this for arguments: -VISION-FACE=False -VISION-SCENE=False -VISION-DETECTION=True -ADMIN-KEY= -API-KEY= -PORT=88
3b. I have this in the IO tab for output and error: C:\DeepStack\Logs\DeepStack.log
3c. Same as 2c
With help from a friend got it to work using this command

docker run -e VISION-DETECTION=True -v localstorage:/datastore -p 83:5000 deepquestai/deepstack
 
With help from a friend got it to work using this command

docker run -e VISION-DETECTION=True -v localstorage:/datastore -p 83:5000 deepquestai/deepstack

Notice any benefit in CPU usage? Is 83 your port (instead of 81)? What is the 5000?
 
1. Yes, i run the services as my normal user
2. I have this in arguments for python: C:\DeepStack\intelligence.py -MODE=Medium -VFACE=False -VSCENE=False -VDETECTION=True
2b. I have this in the IO tab for output and error: C:\DeepStack\Logs\pyIntelligence.log
2c. On File rotation I have enabled: Replace existing files
3. I have this for arguments: -VISION-FACE=False -VISION-SCENE=False -VISION-DETECTION=True -ADMIN-KEY= -API-KEY= -PORT=88
3b. I have this in the IO tab for output and error: C:\DeepStack\Logs\DeepStack.log
3c. Same as 2c

Didnt seem to work at my end.

For this command:

-VISION-FACE=False -VISION-SCENE=False -VISION-DETECTION=True -ADMIN-KEY= -API-KEY= -PORT=88

If your API key was say af-5667-yujhn-ssr and port was 81, by way of example, would it be:

-VISION-FACE=False -VISION-SCENE=False -VISION-DETECTION=True -ADMIN-KEY= -af-5667-yujhn-ssr= -PORT=81

I.e., have a - in front of the actual API key?
 
Failed trying to even install docker desktop for win 10 pro right after clicking on the downloaded installation file: Docker Desktop requires the Server service to be enabled... what service??
 
Failed trying to even install docker desktop for win 10 pro right after clicking on the downloaded installation file: Docker Desktop requires the Server service to be enabled... what service??
I know I had to go into bios and turn on visualization
 
Curious, while not analyzing pictures, how many cams are you running (real and clones) and what is your CPU usage? I'm running this:

1594948521471.png

BI fluctuates quite a bit, from 28% to 40%, up and down. Seems higher than it should be, not sure if the clones should be causing so much usage.
 
Curious, while not analyzing pictures, how many cams are you running (real and clones) and what is your CPU usage? I'm running this:

View attachment 66485

BI fluctuates quite a bit, from 28% to 40%, up and down. Seems higher than it should be, not sure if the clones should be causing so much usage.

13 regular cams and 5 with clones for ai tools
 
I know I had to go into bios and turn on visualization

Visualization? Did you mean virtualization?

If so, mine is already on.
 
13 regular cams and 5 with clones for ai tools

What does BI CPU usage look like on it's own? You mention CPU in total at 70...curious what just BI is.
 
Must still have a setting off somewhere, if I turn off my clones, usage drops to around 18%-22%, but then almost doubles with them on. Didn't think the clones were supposed to take much if any CPU.
 
Must still have a setting off somewhere, if I turn off my clones, usage drops to around 18%-22%, but then almost doubles with them on. Didn't think the clones were supposed to take much if any CPU.
Don't think they should