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

Thank you so much for this guide and tool.
This will stop all of the alerts due to wind and other stupid stuff.

I just wish that there was a way to have DeepStack and the AI Tool to boot up with windows.
My server uses a user without a password, so the above steps to add the service didn't work for me.
But without DeepStack starting too, it's kinda useless anyway.

The only real hurdle that I hit was getting alerts to work via Telegraph; I entered the API token incorrectly :(

You rock, this is amazing.
 
I just wish that there was a way to have DeepStack and the AI Tool to boot up with windows.
My server uses a user without a password, so the above steps to add the service didn't work for me.
But without DeepStack starting too, it's kinda useless anyway.
That should not make a difference.
Services starts whether a user logs in or not.
 
Thank you @Chris Dodge! What happened to GentlePumpkin? is he ok?

I'm totally fine, just haven't got enough time to add new features and work on AI Tool. I'm very happy about the work of @Chris Dodge and ClassObject (and all other contributors too!) and I'm planning to start the next official AI Tool release with the feature base of VorlonCD's fork, but currently I'm completely overloaded with other work, so this might take some time. Many greetings everyone!
 
Thank you so much for this guide and tool.
This will stop all of the alerts due to wind and other stupid stuff.

I just wish that there was a way to have DeepStack and the AI Tool to boot up with windows.
My server uses a user without a password, so the above steps to add the service didn't work for me.
But without DeepStack starting too, it's kinda useless anyway.

The only real hurdle that I hit was getting alerts to work via Telegraph; I entered the API token incorrectly :(

You rock, this is amazing.
VrolonCD integrated a Deepstack control tab into AI Tool which let's you automatically start Deepstack with AI Tool.
 
Just wanted to follow up here to see if your experience running Deepstack on a raspberry pi with an Intel NCS2 has changed after using it for a while? I’m getting ready to pull the trigger on a raspberry pi 4 and a Intel NCS, but wanted to see how it performs when multiple images are sent to it.

Hey mate, very happy with how its running. Haven't seen any processing delays at all. Haven't had to reboot or restart anything, seems very stable.

I think I still need to tweak some of the motion settings in BI for the indoor cameras though, but that is a BI setup issue for me, nothing to do with the Pi.
 
Hey mate, very happy with how its running. Haven't seen any processing delays at all. Haven't had to reboot or restart anything, seems very stable.

I think I still need to tweak some of the motion settings in BI for the indoor cameras though, but that is a BI setup issue for me, nothing to do with the Pi.
Thanks for the update. I'll be placing an order for a Raspberry Pi 4 and a NCS 2 today. Looking forward to testing it out!
 
  • Like
Reactions: Eatoff
Thanks for the update. I'll be placing an order for a Raspberry Pi 4 and a NCS 2 today. Looking forward to testing it out!

I'll be interested to see how you get on as all of mine place the objects in the wrong place, but otherwise they are really good. Hopfully a beta will sort that out.

Pi4 is quite a bit faster than the Pi3B+
 
Is it currently possible for AI Tool, or any of the forks, to save Deepstack's marked-up images (with the object boxes and text labels) to a folder? Ken said it'd be easy enough to use that file as the Alert image in BI, if so.

I spent an hour searching this thread :rolleyes: but the closest thing I could find was using the marked-up images for Telegram.
 
Last edited:
Is it currently possible for AI Tool, or any of the forks, to save Deepstack's marked-up images (with the object boxes and text labels) to a folder? Ken said it'd be easy enough to use that file as the Alert image in BI, if so.

I spent an hour searching this thread :rolleyes: but the closest thing I could find was using the marked-up images for Telegram.
Not yet.
 
  • Like
Reactions: aesterling
Minor update to VorlonCD mod:

  • Ability to specify telegram captions when sending images [ Thank you @mayop ! - @barnyard ]
  • Ability to specify custom filename for image copy action [partial implementation for @aesterling 's request - No saving/merging of markup/detection rectangle yet]
  • Variables added - [Prefix], [ImageFilenameNoExt], [SummaryNonEscaped] (may not always work for filenames and urls)
  • Fix for dynamic mask history sometimes not getting purged

 
Minor update to VorlonCD mod:

  • Ability to specify telegram captions when sending images [ Thank you @mayop ! - @barnyard ]
  • Ability to specify custom filename for image copy action [partial implementation for @aesterling 's request - No saving/merging of markup/detection rectangle yet]
  • Variables added - [Prefix], [ImageFilenameNoExt], [SummaryNonEscaped] (may not always work for filenames and urls)
  • Fix for dynamic mask history sometimes not getting purged

This is greatness! You guys are awesome
 

Attachments

  • Screenshot_20200915-192105.png
    Screenshot_20200915-192105.png
    2 MB · Views: 120
  • screen.png
    screen.png
    24.2 KB · Views: 113
uhh i get mannyy error last 2 versions of vorlonCD ... Spamming Log an Telegram:
[16.09.2020, 08:30:58]: TelegramText>> ERROR: Could not send text "[16.09.2020, 08:30:50]: TelegramText>> ERROR: Could not send text "[16.09.2020, 08:30:50]: TelegramText>> ERROR: Could not send text "[16.09.2020, 08:30:46]: ImageQueueLoop>> * Not showing further warnings about skipping disabled URL: " to Telegram: Response status code does not indicate success: 429 (Too Many Requests). [HttpRequestException] Mod: <TelegramText>d__26 Line:1196:25" to Telegram: Response status code does not indicate success: 429 (Too Many Requests). [HttpRequestException] Mod: <TelegramText>d__26 Line:1196:25" to Telegram: Response status code does not indicate success: 429 (Too Many Requests). [HttpRequestException] Mod: <TelegramText>d__26 Line:1196:25
 
Last edited:
You need the latest VorlonCD version. Then under the camera settings, see my screenshots below. Just customise the text to your liking. There are some variables you can use, as well as just the ability to type in what you want. It's awesome
 

Attachments

  • Capture.PNG
    Capture.PNG
    22.4 KB · Views: 160
  • Capture2.PNG
    Capture2.PNG
    15.5 KB · Views: 156
  • Like
Reactions: balucanb
Can i use the memo / summary /memo / flag to use it on my BI Alert E-Mail ?
I will get a mail: %cam Alert with: person