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

here's whats showing. The last one is what it used to say when I logged in, but I believe it would shut itself down when logged off but I had no way of checking - just that it didn't work at all
 

Attachments

  • Capture5.PNG
    Capture5.PNG
    21.2 KB · Views: 56
  • Capture4.PNG
    Capture4.PNG
    23.2 KB · Views: 54
  • Capture3.PNG
    Capture3.PNG
    18.5 KB · Views: 43
  • Capture2.PNG
    Capture2.PNG
    18.5 KB · Views: 39
  • Capture1.PNG
    Capture1.PNG
    8.5 KB · Views: 41
The initial screenshots look like what I have.

Stop the AlwaysUp service, set it to Disabled, then set the normal Docker service to Disabled, then restart the machine. Log back on again and start Docker Desktop from the icon on your desktop, it should launch.

Then close it, set the normal Docker service to Automatic, click Apply, then Ok then click the little Start link near the top left. Then try launching Docker using the icon on your desktop, you should get a message to say it’s already running.
 
Is there a solution for the Nvidia docker crashing (Image not able to upload image anymore)? Would it be a solution to let docker restart the script every 30min or 1 hour so it keeps running and is this possible?
 
Last edited:
Take the two following url templates, replace {user} and {password} and add both separated by comma to AI Tool:
Code: 
http://localhost:80/admin?camera=[camera]&trigger&user={user}&pw={passwort}
http://localhost:80/admin?camera=[camera]&flagalert=1&trigger&memo=[summary]&user={user}&pw={passwort}
Click to expand...

I'm confused by the instructions. Are both lines required for Trigger URL? Can't I just use the second one? Don't they both trigger and one flags an alert?
 
ChrisX said:
I don't know. Who can I download noavx version for win or only docker?
Click to expand...
The noavx version is only supported in Docker, not Windows natively.

Not sure if you're going to run Docker Desktop or run Docker inside of a Linux box, but for Docker Desktop run the following commands to pull and run the noavx version:

docker pull deepquestai/deepstack:noavx

docker run —name="deepstack" -e TZ=America\New_York -e VISION-DETECTION=True -v localstorage:/datastore -p 80:5000 --restart always deepquestai/deepstack:noavx
 
blulite said:
I'm confused by the instructions. Are both lines required for Trigger URL? Can't I just use the second one? Don't they both trigger and one flags an alert?
Click to expand...
You're right, you don't need both. The second one flags and triggers.
 
  • Like
Reactions: seth-feinberg
IAmATeaf said:
The initial screenshots look like what I have.

Stop the AlwaysUp service, set it to Disabled, then set the normal Docker service to Disabled, then restart the machine. Log back on again and start Docker Desktop from the icon on your desktop, it should launch.

Then close it, set the normal Docker service to Automatic, click Apply, then Ok then click the little Start link near the top left. Then try launching Docker using the icon on your desktop, you should get a message to say it’s already running.
Click to expand...
When following this, upon trying to launch docker desktop manually I get "Service is not running, Docker Desktop Service is not running, would you like to start it? Windows will ask you for elevated access." Same as I get when the service is enabled and set to automatic. I should note that Windows never asks me for elevated access if I click start.

I understand this can work as I just threw docker on another non-BI system with windows 10 and it just worked. I used an old Win 7 Pro key and upgraded my BI PC from Win10 Home to Pro.. nope same issue.. really scratching my head here..

I had to enable virtualization in my BIOS initially, anything else I may need to do that it's not telling me about? I also have Windows 10 set to auto login.. is that the problem??
 
Changed from automatic to automatic delayed start.. seems to be working? Gotta love non-realtime triggered systems. This is why we don't use "time".. race conditions are real people. Especially on repurposed EOL hardware..
 
blazin912 said:
Changed from automatic to automatic delayed start.. seems to be working? Gotta love non-realtime triggered systems. This is why we don't use "time".. race conditions are real people. Especially on repurposed EOL hardware..
Click to expand...
This did not work for me either
 
blulite said:
Thanks for the quick response. But why do the instructions say:
".6 The 'Trigger URL(s)' needs to be filled with these two URLs to trigger and then flag a specific alert in Blue Iris."​
Click to expand...

I’m honestly not sure. @GentlePumpkin may be able to shed some light on the instructions.


Sent from my iPhone using Tapatalk
 
I have to come to terms that my aging pc is not capable of running BI and this workflow. I got it going but any image processing drive the CPU to 100% making the ability to respond to other events unreliable. I'll considering upgrading the box but for the time being I'll deal with my false triggers, more responsive and reliable system... :(
 
blazin912 said:
I have to come to terms that my aging pc is not capable of running BI and this workflow. I got it going but any image processing drive the CPU to 100% making the ability to respond to other events unreliable. I'll considering upgrading the box but for the time being I'll deal with my false triggers, more responsive and reliable system... :(
Click to expand...

What CPU do you have?

With auto logon does that mean that the user has no password?
 
Last edited:
IAmATeaf said:
What CPU do you have?

With auto logon does that mean that the user has no password?
Click to expand...
Auto logon there is a password but it is not required.

CPU is i5 3570 w/ 4GB RAM
CPU normally hovers around 20-40% without this load. RAM 1GB
Maybe the speed is related to my HDDs they are ancient too. This was some old PC that hadn't been used for a few years, I swapped out the CPU from a scavenge seller. All hardware tested through prime95 and stable, but it's definitely slow. Very, very, slow
 
GentlePumpkin said:
In addition to pmcross' answer:

1. Ist BI on port 81?
2. {user} and {pw} need to be replaced, but not [summary]
3. I'm not quite sure, but it could be that there must be at least a few milliseconds between creating a trigger and tagging it with memo=[summary]. So triggering and adding the summary must happen with two subsequent url calls
Click to expand...

Thank you all for the suggestions. I finally got a chance to get back to working on this today - been a busy week!:) I have tried various iterations of the memo variable in the URL to no avail. GentlePumpkin, to answer your questions:

1. Ist BI on port 81? Yes, and Deepstack is on Port 83
2. {user} and {pw} need to be replaced, but not [summary] Confirmed
3. I'm not quite sure, but it could be that there must be at least a few milliseconds between creating a trigger and tagging it with memo=[summary]. So triggering and adding the summary must happen with two subsequent url calls Not sure I followed this. Are you suggesting that the delay between trigger and tag would generate the 2 URL's, or that I NEED to create 2 URL's? How can I go about checking to see if this is happening?

I don't know if this is related to this or not (triggering delays) but it seems that one of my cameras will only randomly generate the alarm clip in BI. It shows up in the AITool history, but I will only get the clip in BI maybe 1 out of 10 times (seems random though). The URL is configured the same as the other cameras, and I keep checking BI to see if something is configured differently for that camera, but I can't seem to see any differences.

Thanks again for everyone's help!
 
blazin912 said:
Auto logon there is a password but it is not required.

CPU is i5 3570 w/ 4GB RAM
CPU normally hovers around 20-40% without this load. RAM 1GB
Maybe the speed is related to my HDDs they are ancient too. This was some old PC that hadn't been used for a few years, I swapped out the CPU from a scavenge seller. All hardware tested through prime95 and stable, but it's definitely slow. Very, very, slow
Click to expand...
Well I had an SSD and 16GB of RAM laying around.. clonezilla'd and dropped it in. All my problems solved. Slow pc = unreliable.

Docker service set to automatic start, no issue, windows 10 pro.. but i suspect would've been fine with home.
 
blazin912 said:
Well I had an SSD and 16GB of RAM laying around.. clonezilla'd and dropped it in. All my problems solved. Slow pc = unreliable.

Docker service set to automatic start, no issue, windows 10 pro.. but i suspect would've been fine with home.
Click to expand...

Was going to say 4Gb of RAM is nothing, Windows was probably swapping like crazy, I tried running my dedicated BI machine with just 8Gb of ram and it was never happy so have 12Gb in it right now, planning to up that to 16Gb at some point which should smooth things out nicely.
 
IAmATeaf said:
Was going to say 4Gb of RAM is nothing, Windows was probably swapping like crazy, I tried running my dedicated BI machine with just 8Gb of ram and it was never happy so have 12Gb in it right now, planning to up that to 16Gb at some point which should smooth things out nicely.
Click to expand...
Yea I was having reliability issues that I think we're linked to two bad ram slots on my mobo. They are intermittent causing no boots periodically. In my debug I swapped sticks around and landed on 2x2gb, but had 2x8gb I forgot about . Anyways this thing flies now.
 
also getting same error as post 1323

docker: Error response from daemon: OCI runtime create failed: container_linux.go:349: starting container process caused "exec: \"for\": executable file not found in $PATH": unknown

followed guide step by step. port 81 was in use so tried 83 and getting the error. on windows 1809 if it makes a difference
 
You must log in or register to reply here.
Top Bottom