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

ReXX · Jul 26, 2020

Syc0Maniac said:
Are those scripts? I have no clue where to start with that or even where to put it :/

How did you make AITool as a service?
Same procedure. I use NSSM.

Also see this post

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

So you've manually added that into your Startup routine via the sc command? https://support.microsoft.com/en-us/help/251192/how-to-create-a-windows-service-by-using-sc-exeActually I used nssm (version 2.24-101-g897c7ad from NSSM - the Non-Sucking Service Manager It also supports redirect...

ipcamtalk.com

Syc0Maniac · Jul 26, 2020

ReXX said:
How did you make AITool as a service?
Same procedure. I use NSSM.

Also see this post

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

So you've manually added that into your Startup routine via the sc command? https://support.microsoft.com/en-us/help/251192/how-to-create-a-windows-service-by-using-sc-exeActually I used nssm (version 2.24-101-g897c7ad from NSSM - the Non-Sucking Service Manager It also supports redirect...

ipcamtalk.com

I do not have python or anything extra besides what I needed to get my BI, AITools and DQ working on a fresh Win 10 Pro install.

ReXX · Jul 26, 2020

Syc0Maniac said:
I do not have python or anything extra besides what I needed to get my BI, AITools and DQ working on a fresh Win 10 Pro install.

Well.. Deepstack is running on python so you have it.

All files is placed in the subfolders of the Deepstack folder.

Syc0Maniac · Jul 26, 2020

ReXX said:
Well.. Deepstack is running on python so you have it.

All files is placed in the subfolders of the Deepstack folder.

So just follow the steps as you did for the AITools but just do those 3 as you posted before instead? I am stuck as what to do @ step 5. Sorry, I am not good with "coding" and follow step/step tutorials till I get the hang of things and this is confusing me where I put this information and or how to even do it. I understand how it was done for AITools but that is the extent of it :/

surfer90 · Jul 27, 2020

Thanks a thousand times for this awesome tool man, I am using it now for about a month and everyone I'm showing it to is completely stoked

Running deepstack on a Pi4 with 4 GB ram and the Intel stick and it works like a charm.
I'm including it in my iobroker home automation to only record videos when both me am the wife are away which is working perfectly. Plus for example start an HD recording when the front gate opens even if there was no person detected.
Start all HD streams when the break in alarm fires etc...
Next step will be pushover notifications with alert images where your newest version with [file name] comes very handy!

One feature request/idea: how about a checkbox to select whether to send telegram alert images with the original image (as it is now) or including the detection boxes overlay? Maybe that's possible already but I wasn't able to find it.

Anyone having questions about the above setup just let me know

Edit: since I was wondering about this before: the Pi setup gives me results in less than 1 second. Even though I'm hitting it with images about 1MB size. Way faster than the 6-10 seconds I was getting with my non avx desktop PC.

ReXX · Jul 27, 2020

Syc0Maniac said:
So just follow the steps as you did for the AITools but just do those 3 as you posted before instead? I am stuck as what to do @ step 5. Sorry, I am not good with "coding" and follow step/step tutorials till I get the hang of things and this is confusing me where I put this information and or how to even do it. I understand how it was done for AITools but that is the extent of it :/

You run the nssm install <service-name>
I called the first one "redis-server" and the config is like this:

Also under the logon tab, enter your username and password. Goes for them all.

The second one I called pyIntelligence, and is dependent on the first one.

Arguments: C:\DeepStack\intelligence.py -MODE=Medium -VFACE=False -VSCENE=False -VDETECTION=True

The last one I called Deepstack-server and is dependent on the second one

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

Syc0Maniac · Jul 27, 2020

ReXX said:
You run the nssm install <service-name>
I called the first one "redis-server" and the config is like this:

View attachment 67269 View attachment 67270
Also under the logon tab, enter your username and password. Goes for them all.

The second one I called pyIntelligence, and is dependent on the first one.
View attachment 67271 View attachment 67272 View attachment 67273
Arguments: C:\DeepStack\intelligence.py -MODE=Medium -VFACE=False -VSCENE=False -VDETECTION=True

The last one I called Deepstack-server and is dependent on the second one
View attachment 67274 View attachment 67275 View attachment 67276
Arguments: -VISION-FACE=False -VISION-SCENE=False -VISION-DETECTION=True -ADMIN-KEY= -API-KEY= -PORT=88

Oh, ok. SO you can call the service anything you want?

ReXX · Jul 27, 2020

Syc0Maniac said:
Oh, ok. SO you can call the service anything you want?

Yes, anything you like..

GentlePumpkin · Jul 27, 2020

surfer90 said:
One feature request/idea: how about a checkbox to select whether to send telegram alert images with the original image (as it is now) or including the detection boxes overlay? Maybe that's possible already but I wasn't able to find it.

Anyone having questions about the above setup just let me know

Yes I can add a switch to turn off sending the image along.

regarding the Raspberry and the Compute stick: what are the processing times each?

GentlePumpkin · Jul 27, 2020

Regarding a Docker GPU version: Using DeepStack with NVIDIA GPUs

it exists, I never tried it out, and if someone did so, please tell us how (fast) ist works!

GentlePumpkin · Jul 27, 2020

new updates including many discussed and additional features and a massively simplified setup are coming in the next month. So stay tuned - and in case you consider setting AI Tool up with many camera, consider waiting. The new setup won‘t require configuring duplicate cams anymore

Senor Pibb · Jul 27, 2020

I am currently running 1.65 of AI Tool. Where can I find 1.67 preview? Thanks

pmcross · Jul 27, 2020

IAmATeaf said:
Did you try the suggestions to install the hyper-v feature and enable virtualisation in the BIOS?

For me, my BI PC is on W10 2004 which now has support for WSL2 (windows subsytem for Linix 2) which is what I opted to use. Did have a right royal struggle to enable virtualisation in the BIOS though as I run headless and the PC is in the loft but eventually found a HP utility that allows you to change BIOS options from the command line, so used that, then restarted, then installed Docker which enabled WSL2 etc....

I have Docker Desktop set up and am trying to get Deepstack to run on it. Can you provide a high level how you set up Docker Desktop? It is different than standard Docker that I'm used to running on Ubuntu. I'm able to do a pull for deepquestai/deepstack, but can't get it to run. I'm getting hung up on the container.

IAmATeaf · Jul 27, 2020

If you look back I posted the command line that I used to do the pull which also starts.

pmcross · Jul 27, 2020

IAmATeaf said:
If you look back I posted the command line that I used to do the pull which also starts.

I got this working now, thanks! Now I will monitor to see how my CPU handles running Deepstack on Docker Desktop rather than the Windows version.

pmcross · Jul 27, 2020

Just wanted to provide a quick update since switching to Docker Desktop. My performance is significantly better. My response times are anywhere from 500ms to just under 1 second now. Also, my CPU hovers around 35 to 40% and spikes to the 70 to 80’s when human activity is present, which is a stark contrast to running Deepstack on Windows, or even running the noavx version on my old PowerEdge 2950

Sent from my iPhone using Tapatalk

Simcole · Jul 27, 2020

Hi all. I just found this project and wow this is awesome. Thanks GentlePumpkin and others for making this so cool.

Is there any other notification possible that's not telegram? I'd to read up on how to get this to send me alerts if someone pulls into my driveway with a picture. I don't have telegram but I do use homeassistant. I also saw that the free version of deepstack recognizes up to 5 faces I believe. Where do you set / train those to get similar alerts if they're strangers?

cjowers · Jul 28, 2020

Hi all, thanks for this, it looks amazing

Unfortunately I get an error after the AiTool sends the image to deepstack server. AiTool log file errors looks like:

(1/6) uploading image
(2/6) waiting for results
Newtonsoft.json.jsonserializationexception |error converting value 404 to type 'windowsformsapp2.response'. path '', line 1, position 3. (code: -2146233088)
error: processing the following image 'xxx.jpg' failed. can't reach deepquest server at http://192.xxx.xx.xx:xx....etc

the deepstack log adds a POST line everytime an image appear in the input folder, when the aitool errors.

Any ideas? Using windows install (not docker) on win10, AItool v1.67. restarted, ran everything as admin, tried troubleshooting (server looks to be running?).

In blue Iris settings - webserver: I configured the advanced options as per the step-by-step instructions, but do i also need to turn on the http server option? and if so, what should the fields / addresses be in that form?
And just to confirm: I don't need to be connected to the internet for this, correct?

Thanks for any help,
C

pmcross · Jul 28, 2020

I’m guessing that your CPU doesn’t support AVX. If it does please refer to post 238 in this thread; I went over the application requirements to run Deepstack on Windows.

Sent from my iPhone using Tapatalk

cjowers · Jul 28, 2020

pmcross said:
I’m guessing that your CPU doesn’t support AVX. If it does please refer to post 238 in this thread; I went over the application requirements to run Deepstack on Windows.

Sent from my iPhone using Tapatalk

Thanks for the help pmcross - I suspect my cpu lacks AVX. its a xeon W2530, probably from 2009ish... Bummer :facepalm:

Any suggestions / options?

edit Looks like post #1100 mentions a non-avx docker version. I suppose I'll try that. Docker Hub

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

Young grasshopper

n3wb

Young grasshopper

n3wb

n3wb

Young grasshopper

n3wb

Young grasshopper

Getting the hang of it

Pulling my weight

Known around here

Pulling my weight

Pulling my weight

n3wb

Getting the hang of it

Pulling my weight

Getting the hang of it