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

Senor Pibb

Young grasshopper
Joined
May 22, 2020
Messages
50
Reaction score
18
Location
Greer, SC
Not sure if you have any sort of home automation, but with Home Assistant I can change profiles based on people being home or not. Works pretty great. Might be able to do it with other home automation software too.
You know I have not done that yet. But have been thinking about that. Wife says I love my projects. In your video you mention turn on lights based on motion. That pretty much move my old butt forward. :) Sometimes I just need a reason to justify it to myself. :) I did install Home Assistant via docker on my UnRaid box. But don't know the default user name and p/w . I did try the normal admin/admin or admin/password. If anyone else has downloaded HA docker from homeassistant/home-assistant Repository. let me know what the login info is plz.
 

Chris Dodge

Pulling my weight
Joined
Aug 9, 2019
Messages
78
Reaction score
102
Location
massachusetts
Hi Folks. Here is the latest update to the VorlonCD mod. Should address many of the issues or suggestions mentioned recently.


  • Splash screen and much faster startup. The database cleanup of missing image files happens on startup, but it generally doesn't take too long.
  • @classObject has done some great Dynamic Masking updates! A few bug fixes, and an advanced setting "Enable Object Scaling" that gives the ability to scale distant objects so they are more easily detected as the same object that may have been detected before. (Without this setting, you may have found you had to create half a dozen static dynamic masks for a distant object you wanted to ignore)
    image
    image
  • Redesigned Log Viewer. Faster, Filter on keywords or errors, sort by original log order or date, etc. Log files will be deleted when over the "MaxLogFileAgeDays" (defaults to 14) setting in the AITOOL.SETTINGS.JSON file, rotated out when the size is over "MaxLogFileSize" (bytes - defaults to 10MB), and ZIPPED up once a day.
    image
  • Redesigned History Viewer. Better icons, all settings get saved, prediction details screen, jump directly to mask settings, Follow history option works better.
    image
  • History - Right-click menu. Locate in log file, refresh, etc.
    image
  • History > Prediction Details > right-click > Create static mask. So you can create static masks after they have been cleared in the Dynamic Mask Details screen. Note this works with multiple items selected.
  • History > Settings: Adds ability to prevent false or masked history items from being stored in the database. In some cases this may make list loading and filtering faster but probably not necessary in most cases.
  • History > Settings: Enable "Show only Relevant objects" to closer mimic @GentlePumpkin 's original app where in the history list, highlighted objects and variables only showed relevant objects. With it unchecked you get a lot more insight for what deepstack is outputting.
  • When more than one object is detected, the list is now prioritized so that relevant objects will be first in any list. Then other 'Important' objects like 'person', 'bear' will be prioritized. [variables] will now show relevant objects first. This can be controlled via the ObjectPriority setting in the JSON settings file. View the detected object list via History > Prediction Details
    image
  • Settings now stored under _SETTINGS folder and logs are under _SETTINGS\LOGS folder so everything that changes is stored under a single folder. CAMERAS folder is no longer used. I know, I said that before, but this time its true. Really. Trust me?
  • Camera > Apply To button. This lets you save selected settings from the current camera to any other camera.
  • Actions - "Queue Actions" checkbox - This will send actions that need to be done like sending Telegram, MQTT or calling a trigger URL into a background thread queue to be processed. It will release the main detection routine faster so you dont have to wait for a telegram upload to finish before processing the next image from BI. Note that this may affect Cooldown behavior.
  • Actions - New variable [ImagePathEscaped]. This allows you to tell BlueIris which image to display in its alert list using the following url parameter: &flagalert=1&memo=[summary]&jpeg=[ImagePathEscaped] (Note you must have a recent version of BI for this to work)
  • Actions - Ability to change JPEG compression level when "Merge Annotations into Images" is checked.
  • Actions - Telegram: Ability to specify which objects work work with - for example, "person, dog, bear" Leave empty for all.
    image
  • Fix: Cancel triggers now working as expected.
  • Fix: Switching to high-dpi support caused all sorts of issues for people remoting into a machine with AITOOL so that has been disabled. Fuck me, why did I buy a 4k display? :)
  • Documentation on how to setup MQTT with BlueIris thanks to @edalquist!: VorlonCD/bi-aidetection
  • Many other fixes. Github issues is best way log new issues, but do log to the ipcamtalk thread also since I dont always monitor both. Jeebers, 114 frikin pages! :) https://github.com/VorlonCD/bi-aidetection/issue

===================
To install:


Backup your current AITOOLS folder.
Download the ZIP, right-click > Properties > UNBLOCK.
Unzip to existing folder, overwriting any files.
Most if not all settings should migrate.

OH, and holey crap, thanks to @Tinbum, @aesterling, @petermai6655, @aralos1999 among many others for their suggestions, bug reports and patience.
 
Last edited:

105437

BIT Beta Team
Joined
Jun 8, 2015
Messages
1,070
Reaction score
238
Many thanks to GP for the original code, and CD for the awesome additions.

All I need now is Deepstack to release a Jetson Nano version and I'm golden!
And a Windows 10 version of the new release! Would also like to see additional objects/animals added to Deepstack.
 

tripp396

Getting the hang of it
Joined
Jun 18, 2020
Messages
65
Reaction score
30
Location
Minnesota
Hi Folks. Here is the latest update to the VorlonCD mod. Should address many of the issues or suggestions mentioned recently.


  • Splash screen and much faster startup. The database cleanup of missing image files happens on startup, but it generally doesn't take too long.
  • @classObject has done some great Dynamic Masking updates! A few bug fixes, and an advanced setting "Enable Object Scaling" that gives the ability to scale distant objects so they are more easily detected as the same object that may have been detected before. (Without this setting, you may have found you had to create half a dozen static dynamic masks for a distant object you wanted to ignore)
    image
    image
  • Redesigned Log Viewer. Faster, Filter on keywords or errors, sort by original log order or date, etc. Log files will be deleted when over the "MaxLogFileAgeDays" (defaults to 14) setting in the AITOOL.SETTINGS.JSON file, rotated out when the size is over "MaxLogFileSize" (bytes - defaults to 10MB), and ZIPPED up once a day.
    image
  • Redesigned History Viewer. Better icons, all settings get saved, prediction details screen, jump directly to mask settings, Follow history option works better.
    image
  • History - Right-click menu. Locate in log file, refresh, etc.
    image
  • History > Prediction Details > right-click > Create static mask. So you can create static masks after they have been cleared in the Dynamic Mask Details screen. Note this works with multiple items selected.
  • History > Settings: Adds ability to prevent false or masked history items from being stored in the database. In some cases this may make list loading and filtering faster but probably not necessary in most cases.
  • History > Settings: Enable "Show only Relevant objects" to closer mimic @GentlePumpkin 's original app where in the history list, highlighted objects and variables only showed relevant objects. With it unchecked you get a lot more insight for what deepstack is outputting.
  • When more than one object is detected, the list is now prioritized so that relevant objects will be first in any list. Then other 'Important' objects like 'person', 'bear' will be prioritized. [variables] will now show relevant objects first. This can be controlled via the ObjectPriority setting in the JSON settings file. View the detected object list via History > Prediction Details
    image
  • Settings now stored under _SETTINGS folder and logs are under _SETTINGS\LOGS folder so everything that changes is stored under a single folder. CAMERAS folder is no longer used. I know, I said that before, but this time its true. Really. Trust me?
  • Camera > Apply To button. This lets you save selected settings from the current camera to any other camera.
  • Actions - "Queue Actions" checkbox - This will send actions that need to be done like sending Telegram, MQTT or calling a trigger URL into a background thread queue to be processed. It will release the main detection routine faster so you dont have to wait for a telegram upload to finish before processing the next image from BI. Note that this may affect Cooldown behavior.
  • Actions - New variable [ImagePathEscaped]. This allows you to tell BlueIris which image to display in its alert list using the following url parameter: &flagalert=1&memo=[summary]&jpeg=[ImagePathEscaped] (Note you must have a recent version of BI for this to work)
  • Actions - Ability to change JPEG compression level when "Merge Annotations into Images" is checked.
  • Actions - Telegram: Ability to specify which objects work work with - for example, "person, dog, bear" Leave empty for all.
    image
  • Fix: Cancel triggers now working as expected.
  • Fix: Switching to high-dpi support caused all sorts of issues for people remoting into a machine with AITOOL so that has been disabled. Fuck me, why did I buy a 4k display? :)
  • Documentation on how to setup MQTT with BlueIris thanks to @edalquist!: VorlonCD/bi-aidetection
  • Many other fixes. Github issues is best way log new issues, but do log to the ipcamtalk thread also since I dont always monitor both. Jeebers, 114 frikin pages! :) https://github.com/VorlonCD/bi-aidetection/issue

===================
To install:


Backup your current AITOOLS folder.
Download the ZIP, right-click > Properties > UNBLOCK.
Unzip to existing folder, overwriting any files.
Most if not all settings should migrate.

OH, and holey crap, thanks to @Tinbum, @aesterling, @petermai6655, @aralos1999 among many others for their suggestions, bug reports and patience.
Is there documentation around how cancel triggers work? Is it in blue iris documentation?
 

balucanb

Getting the hang of it
Joined
Sep 19, 2020
Messages
141
Reaction score
23
Location
TX
Probably nothing but-

I just installed the very latest (from the post above) I like to start clean as as I can, so I delete any all pictures in my AI folder, clear out all the snaps shots, video's, errors etc. from BI before I install a new version. I removed the Cameras folder as directed and then unzipped everything into the existing folder (after backing up) On start up I am getting several hundred of these errors, the errors all differ a bit ref. info on the line but this is a good example of what I see :

DeepStack>> 2020-10-31 09:35:11.1455872 [W:eek:nnxruntime:InferenceSession, session_state_initializer.cc:500 onnxruntime::SaveInputOutputNamesToNodeMapping] Graph input with name 294 is not associated with a node.

So far seems to be running OK?, it appears to sort itself out in a few seconds and I normally won't see this anymore.

So Question is- Normal? or am I jacking it up somehow?

TIA- Love you guys!
 

Chris Dodge

Pulling my weight
Joined
Aug 9, 2019
Messages
78
Reaction score
102
Location
massachusetts
@balucanb yeah I forgot to exclude deepstack (for windows) debug messages. They are nothing to worry about and always happen. Try the DOCKER version rather than windows. It has a newer version. I think instructions may be at top of this thread and mentioned a few other places in the last few 1000 pages here.

@tripp396 - See this: - page 186
 

Netscan

n3wb
Joined
Oct 9, 2019
Messages
8
Reaction score
2
Location
Florida, USA
I'm getting constant external triggers now from "irrelevant" detections - if there's motion in a scene, AI and deepstack do their thing and determine it's a false alarm, but AIT is triggering Blueiris for some reason - like every second until there's an actual alert or i restart AIT. Think it has something to do with the cancel string?
I reverted to the previous version i was running (1.67.8.33855) and it's back to working correctly
 

Chris Dodge

Pulling my weight
Joined
Aug 9, 2019
Messages
78
Reaction score
102
Location
massachusetts
@Netscan you didnt give much info about your configuration. How is BI setup, what is your trigger/cancel url set to? What are the log entries? Are you on the latest version of BlueIris? Flagalert=0 has only been in recent versions.

I believe this is the correct cancel URL:

I dont believe the '?trigger" should be in the cancel URL

Can anyone else verify cancel alerts are working the the 2.0 release, and what the correct URL is?
 

aesterling

Pulling my weight
Joined
Oct 9, 2017
Messages
219
Reaction score
148
@Netscan you didnt give much info about your configuration. How is BI setup, what is your trigger/cancel url set to? What are the log entries? Are you on the latest version of BlueIris? Flagalert=0 has only been in recent versions.

I believe this is the correct cancel URL:
flagalert=0

I dont believe the '?trigger" should be in the cancel URL

Can anyone else verify cancel alerts are working the the 2.0 release, and what the correct URL is?
I'm using cancel URL:


and it works well.
 

Netscan

n3wb
Joined
Oct 9, 2019
Messages
8
Reaction score
2
Location
Florida, USA
@Netscan you didnt give much info about your configuration. How is BI setup, what is your trigger/cancel url set to? What are the log entries? Are you on the latest version of BlueIris? Flagalert=0 has only been in recent versions.

I believe this is the correct cancel URL:

I dont believe the '?trigger" should be in the cancel URL

Can anyone else verify cancel alerts are working the the 2.0 release, and what the correct URL is?
OK, i did have trigger in the cancel URL, i've taken it out and it seems to have fixed it, thanks! Weird how it was working in the other version, huh?
 
Joined
Sep 21, 2020
Messages
8
Reaction score
6
Location
UK
Can anyone help me with a problem I'm having? With the latest VorlonCd release 2.0 and one I compiled myself a couple of weeks ago, I seem to have lost the caption on Telegram messages. I don't think I've changed any settings related to this but I'm attaching an example screen capture of one of the cameras setting page. Even if I replace the Caption setting with simple text nothing is displayed in Telegram except the picture. I've already tried completely removing and reinstalling the tool. Although I expected to have to add all the cameras back in they were restored from the registry. There's nothing I can see in the log file either.
Untitled.png
 

Village Guy

Getting the hang of it
Joined
May 6, 2020
Messages
210
Reaction score
99
Location
UK
Can anyone help me with a problem I'm having? With the latest VorlonCd release 2.0 and one I compiled myself a couple of weeks ago, I seem to have lost the caption on Telegram messages. I don't think I've changed any settings related to this but I'm attaching an example screen capture of one of the cameras setting page. Even if I replace the Caption setting with simple text nothing is displayed in Telegram except the picture. I've already tried completely removing and reinstalling the tool. Although I expected to have to add all the cameras back in they were restored from the registry. There's nothing I can see in the log file either.
View attachment 73907
Have you tried installing the latest version released yesterday. This is a fast moving project and is better supported when you are running the latest and greatest revision.
 

105437

BIT Beta Team
Joined
Jun 8, 2015
Messages
1,070
Reaction score
238
I have a QNAP storage array. I'd like to install Deepstack on the QNAP using the Container Station application on the QNAP. I don't have any experience with that app yet but what kind of virtual OS should I create for Deepstack?
 
Top