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

davidtja

n3wb
Joined
Nov 26, 2020
Messages
3
Reaction score
0
Location
London
I'm a newbie so please be patient. I have recently purchased BI & a new Windows PC to run this excellent set up on. I've tried setting up following both different YouTube install methods by The Hook Up" (using High Def & Low def) and "FamilyTechExec" (using two streams on one camera). I cant seem to get either working. In each case I get loads of unwanted clips - I assume in my alerts I should have ONLY clips that have been processed by AI and have passed as identified movement. AI Seems to work and successfully processes the events but they dont seem to be going through to BI. I have tested the URL by copying into the browser and it gives back the correct response.
I have spent maybe ten hours on this already and wondered if there's a guide anywhere else based on the latest AI. I have run it all under Windows but not adverse to setting up Docker/VM (again i'd need to follow a guide). Any help advice appreciated,
Thanks for bearing with me.
 

Village Guy

Pulling my weight
Joined
May 6, 2020
Messages
223
Reaction score
100
Location
UK
I'm a newbie so please be patient. I have recently purchased BI & a new Windows PC to run this excellent set up on. I've tried setting up following both different YouTube install methods by The Hook Up" (using High Def & Low def) and "FamilyTechExec" (using two streams on one camera). I cant seem to get either working. In each case I get loads of unwanted clips - I assume in my alerts I should have ONLY clips that have been processed by AI and have passed as identified movement. AI Seems to work and successfully processes the events but they dont seem to be going through to BI. I have tested the URL by copying into the browser and it gives back the correct response.
I have spent maybe ten hours on this already and wondered if there's a guide anywhere else based on the latest AI. I have run it all under Windows but not adverse to setting up Docker/VM (again i'd need to follow a guide). Any help advice appreciated,
Thanks for bearing with me.
Most if not all you need to know is within this forum thread. I understand that it's a hell of a lot to read through but to give you a starting point, please read my posts over the last few weeks and then we can try and fill in the blanks ;)
 

Denizz

n3wb
Joined
Dec 9, 2020
Messages
3
Reaction score
0
Location
NL
@Chris Dodge I'm trying out Amazon Rekognition tonight. I have it all set up and working. Is there a way for AI Tool to support the various labels? I'd like to add "Deer" so I get alerts when they trigger an alert. Is that what "Additional Relevant Objects" is for? Can I just add Deer there and have it trigger the camera to record on that camera? Thanks!
I too would like to know about the above question, can I use my own labels / a own trained model with this? Sorry I have not heared about BI until today.
 
Joined
Nov 17, 2020
Messages
16
Reaction score
8
Location
USA
I too would like to know about the above question, can I use my own labels / a own trained model with this? Sorry I have not heared about BI until today.
The additional relevant objects field in AITool can for sure be used to tell AITool to report objects that can be detected by deepstack but aren't available as one of AITool's checkboxes (Deepstack can detect around 80 objects but AITool only has checkboxes for 15 of them). I do not know if/how the entries in this field can generalize to other image analytics software.
 
Joined
Nov 17, 2020
Messages
16
Reaction score
8
Location
USA
Setup a profile in BI that disables detections and then send a command to BI to switch profiles from your phone to activate or de-activate.
I run Tasker on my phone (android) and have it programmed to turn on detection when my phone is out of range of my WiFi signal and off again when I'm home.
Bit like geofencing except based on my wifi signal.
If you simply want to do it manually it is even easier, just send command lines in the same way you have AITool send commands to BI.
Needless to say you will need to open a port on your router to pass through your commands.
I just want to follow-up in case somebody else wants to do this.

I made 2 additional profiles, one for 30 minutes and one for 3 hours. I set the temp time for each to 30 and 180 minutes respectively. The only change for each of these profiles is for each cam I unchecked the box to dump JPEGS when triggered (there are other ways to do it as well). I then installed an app on my phone called "HTTP Shortcuts" and created 3 shortcuts on my home screen, one for each of the new profiles and one for the default profile in case I want to "cancel" the detection pause request. Note that I don't otherwise use profiles or schedules, I just have profile 1 scheduled all the time as is default. You might have to do some more tweaking to this method if you have a schedule that changes profiles periodically already.

The http request for profile change is:

http://192.168.0.xx:xx/admin?user=user&pw=password&profile=x (replace your own IP, user, pw, and profile number)

No need for port forwarding as long as my phone is connected to home wifi or my VPN server.

It works exactly as desired, one single tap on my home screen and AI detections are paused until time runs out or I manually cancel the pause.
 

Village Guy

Pulling my weight
Joined
May 6, 2020
Messages
223
Reaction score
100
Location
UK
I just want to follow-up in case somebody else wants to do this.

I made 2 additional profiles, one for 30 minutes and one for 3 hours. I set the temp time for each to 30 and 180 minutes respectively. The only change for each of these profiles is for each cam I unchecked the box to dump JPEGS when triggered (there are other ways to do it as well). I then installed an app on my phone called "HTTP Shortcuts" and created 3 shortcuts on my home screen, one for each of the new profiles and one for the default profile in case I want to "cancel" the detection pause request. Note that I don't otherwise use profiles or schedules, I just have profile 1 scheduled all the time as is default. You might have to do some more tweaking to this method if you have a schedule that changes profiles periodically already.

The http request for profile change is:

http://192.168.0.xx:xx/admin?user=user&pw=password&profile=x (replace your own IP, user, pw, and profile number)

No need for port forwarding as long as my phone is connected to home wifi or my VPN server.

It works exactly as desired, one single tap on my home screen and AI detections are paused until time runs out or I manually cancel the pause.
Your welcome!
 
Joined
Nov 17, 2020
Messages
16
Reaction score
8
Location
USA
Has anybody else noticed that the x5-beta version of deepstack, while faster, is not as good as the x3 version when it comes to quality of detections? I'm getting more phantom people detected in my images than I used to. I also notice that the people that are detected come with much lower confidence percentages. A person clearly standing in view looking at the camera might still only come back is "65%" whereas on x3-beta it would definitely be 99% or 100%.

It hasn't grossly affecting the function of my system yet but I do get more false positives here and there.
 

cscoppa

Getting the hang of it
Joined
Dec 14, 2019
Messages
48
Reaction score
25
Has anybody else noticed that the x5-beta version of deepstack, while faster, is not as good as the x3 version when it comes to quality of detections? I'm getting more phantom people detected in my images than I used to. I also notice that the people that are detected come with much lower confidence percentages. A person clearly standing in view looking at the camera might still only come back is "65%" whereas on x3-beta it would definitely be 99% or 100%.

It hasn't grossly affecting the function of my system yet but I do get more false positives here and there.
I'm seeing this also. I had a potted plant detected as a human, also confidence percentages on humans is definitely lower.
 
Joined
Nov 17, 2020
Messages
16
Reaction score
8
Location
USA
I'm seeing this also. I had a potted plant detected as a human, also confidence percentages on humans is definitely lower.
Funny, I'm actually seeing the exact same false label. x3 always correctly labelled a "Potted Plant" on my doorstep. x5 detects the same thing as a "Person". Frustrating because it's not in an area I can mask without literally masking half my doorstep.

I might try changing the mode to "High" when I get a chance although I'm not sure how to do that with the Docker version of deepstack.
 

cscoppa

Getting the hang of it
Joined
Dec 14, 2019
Messages
48
Reaction score
25
Funny, I'm actually seeing the exact same false label. x3 always correctly labelled a "Potted Plant" on my doorstep. x5 detects the same thing as a "Person". Frustrating because it's not in an area I can mask without literally masking half my doorstep.
Exact same thing, x3 always identified the plant correctly. x5 sometimes gets it right, sometimes doesn't.
 

105437

BIT Beta Team
Joined
Jun 8, 2015
Messages
1,215
Reaction score
327
I encourage everyone to try AWS Rekognition for image analysis. It's working very well for me.
 

pmcross

Pulling my weight
Joined
Jan 16, 2017
Messages
370
Reaction score
184
Location
Pennsylvania
I encourage everyone to try AWS Rekognition for image analysis. It's working very well for me.
Can you elaborate on your setup? Are you using AI Tool to send images to Rekognition for analysis?


Sent from my iPhone using Tapatalk
 

105437

BIT Beta Team
Joined
Jun 8, 2015
Messages
1,215
Reaction score
327
Can you elaborate on your setup? Are you using AI Tool to send images to Rekognition for analysis?


Sent from my iPhone using Tapatalk
Yes, the version of AI Tool I'm running has Amazon Rekognition as an option. It's the 11/30/20 release. I have 12 Mbps upload speed and my total time for each image is between 700 - 1,500 milliseconds.

Release Notes:
Amazon Rekognition support. Seems to have really good detection ability so far. Enter 'amazon' in Settings > AI Server setting. It has 5000 free images a month, but you must first set up an account, give credit card info AND download the ROOTKEY.CSV file and place in _SETTINGS folder after creating a new user: Step 1: Set up an AWS account and create an IAM user - Amazon Rekognition

I intend on adding the ability to resize images and jpeg quality before sending to Rekognition to reduce the filesize that has to be transmitted. Which should increase the speed.
 
Joined
Nov 17, 2020
Messages
16
Reaction score
8
Location
USA
There's no doubt in my mind that AWS Rekognition is going to give amazing results, and I would love to try it, but...... personally a non-local solution is a bit of a non-starter for me, I've tried hard to make this a fully functioning local system and I don't really want to stray away from that now. Plus I live on a busy downtown street and currently deepstack is processing over 1,000 images per day, and that's AFTER spending the last 4+ months doing as much motion detection optimization as I can. That's well outside the AWS free tier and would have me spending almost $400 per year in fees! Deepstack it is, for now.

I'd love to try other local image analysis options and would be absolutely willing to pay for a stellar one, but I don't know of any others.... are there even any?
 

105437

BIT Beta Team
Joined
Jun 8, 2015
Messages
1,215
Reaction score
327
There's no doubt in my mind that AWS Rekognition is going to give amazing results, and I would love to try it, but...... personally a non-local solution is a bit of a non-starter for me, I've tried hard to make this a fully functioning local system and I don't really want to stray away from that now. Plus I live on a busy downtown street and currently deepstack is processing over 1,000 images per day, and that's AFTER spending the last 4+ months doing as much motion detection optimization as I can. That's well outside the AWS free tier and would have me spending almost $400 per year in fees! Deepstack it is, for now.

I'd love to try other local image analysis options and would be absolutely willing to pay for a stellar one, but I don't know of any others.... are there even any?
Completely see your use case. I live in a rural area so the number of images processed is not a real concern for me. Previously I was running Deepstack via a Docker container on my QNAP. Worked well but I wanted to try Rekognition and see if it could identify more wildlife.
 

pmcross

Pulling my weight
Joined
Jan 16, 2017
Messages
370
Reaction score
184
Location
Pennsylvania
Yes, the version of AI Tool I'm running has Amazon Rekognition as an option. It's the 11/30/20 release. I have 12 Mbps upload speed and my total time for each image is between 700 - 1,500 milliseconds.

Release Notes:
Amazon Rekognition support. Seems to have really good detection ability so far. Enter 'amazon' in Settings > AI Server setting. It has 5000 free images a month, but you must first set up an account, give credit card info AND download the ROOTKEY.CSV file and place in _SETTINGS folder after creating a new user: Step 1: Set up an AWS account and create an IAM user - Amazon Rekognition

I intend on adding the ability to resize images and jpeg quality before sending to Rekognition to reduce the filesize that has to be transmitted. Which should increase the speed.
Thanks for this. I am going to give it a try. I agree with others about the cost of Rekonition due to exceeding 500 images per month. I’m not exactly sure how many images Deepstack processes daily for me, but I imagine that I exceed 500 images a month.


Sent from my iPhone using Tapatalk
 

105437

BIT Beta Team
Joined
Jun 8, 2015
Messages
1,215
Reaction score
327
Thanks for this. I am going to give it a try. I agree with others about the cost of Rekonition due to exceeding 500 images per month. I’m not exactly sure how many images Deepstack processes daily for me, but I imagine that I exceed 500 images a month.


Sent from my iPhone using Tapatalk
It's 5,000 images per month.
 

djborden

Young grasshopper
Joined
Jul 19, 2017
Messages
32
Reaction score
5
Thanks all. What finally sorted it out for me was to reinstall and then run deepquestai/deepstack:noavx using powershell. that way i could specify the port number of the deepstack url, and the VISION-DETECTION=True (not scene-detection) [code shown below], and then I used that port info for the AITool settings tab Deepstack URL.

PS C:\Users\me> docker run -e VISION-DETECTION=True -v localstorage:/datastore -p 85:5000 deepquestai/deepstack:noavx

then the AITool setting tab deepstack URL used was: :85

and what was probably obvious for everyone, but I never found any info on... BI - settings - webserver tab - http server option - should be ticked 'on'. (and use same LAN and port # that you use in AITool camera tab - Trigger URL field)

Thanks pmcross, gentle pumpkin, and any other content creators!
Ive spent hours trying to get Deepstack working in Docker Desktop on Windows 10. I got it all working using the Deepstack Windows 10 install, but based on everyones suggestions here, it made sense to get it running in Docker.

First I had issues figuring out how to manually set the ports, then once I figured that out, it still would not work. Once I downloaded the image to Docker, running this line from the Windows Power Shell and changing the port in AI Tools to 85, it worked !! I have the AVX version so left that off your command line.

-------------------------
C:\Users\me> docker run -e VISION-DETECTION=True -v localstorage:/datastore -p 85:5000 deepquestai/deepstack
------------------------

Will it autostart automatically if the machine reboots?

Thank you, again!
 
Top