[tool] [tutorial] Free AI Person Detection for Blue Iris
- Thread starter GentlePumpkin
- Start date
Has anybody found that AITool will randomly wipe the detection mask for one camera for seemingly no reason? I can't figure out what's triggering it but it's happened 2 or 3 times in recent months, the mask for one of my cams will just be gone and I'll suddenly start getting telegram alerts from areas I don't care about. I think the last time it happened on a Docker update but that doesn't really make sense, not sure how they would be connected like that.
Edit: Never mind, AITool didn't clear the image mask, Blue Iris just started saving substream snapshots instead of main stream snapshots making the mask the wrong size and invisible. Anybody having this problem with the latest update, BI no longer saving main stream snapshots?
Edit: Never mind, AITool didn't clear the image mask, Blue Iris just started saving substream snapshots instead of main stream snapshots making the mask the wrong size and invisible. Anybody having this problem with the latest update, BI no longer saving main stream snapshots?
Last edited:
Had a similar issues so forced the size - there is a tick on the video tab (anamorphic (force size) and made sure I only had either the sub or main stream in the video configuration. My clones used for continuous recording for example have the sub stream string in the field for main and sub empty. This way it is forced to use the sub and nothing else.
im getting a issue,
Give the current user (DESKTOP-CU6RNTD\XXXXXXX) access to '192.168.1.8': Attempted to perform an unauthorized operation. [UnauthorizedAccessException] (In BlueIris.RefreshBIInfoAsync:130 > BlueIris<>c__DisplayClass15_0<RefreshBIInfoAsync>b__0:130 > Win32.RegistryKey.OpenRemoteBaseKey(RegistryHive hKey, String machineName, RegistryView view) > Win32.RegistryKey.Win32ErrorStatic(Int32 errorCode, String str))
how to give permission ?
also i'm not using the integrated version of blueiris as it can not queue images at once, it does only 1 image at the time..
Give the current user (DESKTOP-CU6RNTD\XXXXXXX) access to '192.168.1.8': Attempted to perform an unauthorized operation. [UnauthorizedAccessException] (In BlueIris.RefreshBIInfoAsync:130 > BlueIris<>c__DisplayClass15_0<RefreshBIInfoAsync>b__0:130 > Win32.RegistryKey.OpenRemoteBaseKey(RegistryHive hKey, String machineName, RegistryView view) > Win32.RegistryKey.Win32ErrorStatic(Int32 errorCode, String str))
how to give permission ?
also i'm not using the integrated version of blueiris as it can not queue images at once, it does only 1 image at the time..
whoami ™
Pulling my weight
GreyTheOld
Young grasshopper
- Oct 26, 2017
- 40
- 6
Adding this just in case anyone runs into the same issues, I've been chasing my tail for a few weeks now and JUST solved it.
My setup is 11 cams of various makes and models, but I've had a terrible time with detection working for my Annke I51DM (C500), it all tests right but doesn't reliably trigger, although testing in AITool it works every time. Every time I turned on a detection zone it would stop working. I read someone else's comment about AiTool's dynamic mask looking at a reference image that was the wrong size and that gave me an idea.
First off, the substream URL of these camera's is (imo) hard to find. Here's the substream URL for an Annke I51DM:
192.168.x.xxx/h265/ch1/sub/av_stream
What I noticed in my substream camera setup is that the resolution was still set to 2560x1440 even though the stream profile on the cam is set to 640x480, and I couldn't change it. Changing it to anamamorphic and 640x480, it immediately started working. Not sure why it wasn't detecting correctly, or maybe it's trying to detect the main stream resolution.
My setup is 11 cams of various makes and models, but I've had a terrible time with detection working for my Annke I51DM (C500), it all tests right but doesn't reliably trigger, although testing in AITool it works every time. Every time I turned on a detection zone it would stop working. I read someone else's comment about AiTool's dynamic mask looking at a reference image that was the wrong size and that gave me an idea.
First off, the substream URL of these camera's is (imo) hard to find. Here's the substream URL for an Annke I51DM:
192.168.x.xxx/h265/ch1/sub/av_stream
What I noticed in my substream camera setup is that the resolution was still set to 2560x1440 even though the stream profile on the cam is set to 640x480, and I couldn't change it. Changing it to anamamorphic and 640x480, it immediately started working. Not sure why it wasn't detecting correctly, or maybe it's trying to detect the main stream resolution.
Last edited:
I've tried updating to the latest a couple of times but keep getting these errors:
'Deepstack v3.4 not supported. Install version 2020. Using DeepStack with Windows 10 (CPU and GPU) | DeepStack v1.2.1 documentation'
I use the latest deepstack on docker desktop as a container.
deepquestai/deepstack:latest
Should I be using a different version?
Perhaps:
deepquestai/deepstack:cpu
Do I need to choose the 'Full installation' instead of 'keep of Just extract/update files' and load into program files instead? This will surely lose all my previous configs though.
For now uninstalled and continue using version 2.0.1037.7763
Until I know what's going on with the new version and deepstack in docker.
Hopefully someone here will shed light on this error I'm getting
'Deepstack v3.4 not supported. Install version 2020. Using DeepStack with Windows 10 (CPU and GPU) | DeepStack v1.2.1 documentation'
I use the latest deepstack on docker desktop as a container.
deepquestai/deepstack:latest
Should I be using a different version?
Perhaps:
deepquestai/deepstack:cpu
Do I need to choose the 'Full installation' instead of 'keep of Just extract/update files' and load into program files instead? This will surely lose all my previous configs though.
For now uninstalled and continue using version 2.0.1037.7763
Until I know what's going on with the new version and deepstack in docker.
Hopefully someone here will shed light on this error I'm getting
Last edited:
Update:
What's actually happening is the "flagged" event that shows up in Blue Iris is one flagged event older than the event that triggered it. I realize that's hard to follow.
AI hits on a human at 7 PM. I immediately get a correct telegram alert with the correct picture, but the event doesn't show up in Flagged clips in Blue Iris.
AI hits on another human at 8 PM. I once again get a correct telegram alert and picture. At this point, the 7 PM AI hit now shows up in flagged clips in Blue Iris.
It's like AITool is flagging the previous event with each hit instead of the current. I'm not sure if the problem is with AITool or Blue Iris. Any ideas?
Last edited:
What version of BI are you running, just curious...
Blue Iris Updates (64 bit) - Updates
5.5.1.6, the latest.
This has been happening for awhile, at least the last couple months. I usually always go to the latest update within a couple days of release.
I experimented with using windows deepstack last nigth wih the AI tool.
The first observation is that I could not set Mode=medium on a custom model and High on the standard model. Had to be the same for both.
2nd: CPU spiked much more than running deepstack in docker. There I have the cpu cores limited so it's not unlimited. Even with this set the images are processed at the same speed, around 350ms.
Also, my issues still remain where I cannot install the latest AI version with the errors showing noted in post #3532.
Back to docker on deepstack it was. Was a disappointent as I wanted to shut down docker to reduce the memory used by the Vmmem process.
I'm trying out deepstack on BI today, not sure how the AI works yet without Alerts enabled - maybe not possible at all. I need to read through the notes and posts
The first observation is that I could not set Mode=medium on a custom model and High on the standard model. Had to be the same for both.
2nd: CPU spiked much more than running deepstack in docker. There I have the cpu cores limited so it's not unlimited. Even with this set the images are processed at the same speed, around 350ms.
Also, my issues still remain where I cannot install the latest AI version with the errors showing noted in post #3532.
Back to docker on deepstack it was. Was a disappointent as I wanted to shut down docker to reduce the memory used by the Vmmem process.
I'm trying out deepstack on BI today, not sure how the AI works yet without Alerts enabled - maybe not possible at all. I need to read through the notes and posts
Thank you this is occurring to me too
Is AI Tool still in development? Its a great too and use it instead of BI.
I'm not sure. I come back to this thread every once in awhile but keeping up with the status of the tool is hard for me, I don't have time to read through 100+ pages every time I need to re-acquaint myself with the status/current version. Not bashing, I'm very grateful for the tool. I'd happily pay $$ for a more refined user experience. Maybe there is a more trimmed down status page I don't know about?
I have 2 questions:
1) I installed everything again. (8 instances of Deepstack in Docker + settings in AItool) Same computer and same settings as previous installation. Somehow VMMEM consumes 90% of CPU of a Ryzen 5950X at 4.5Ghz. Even 5-10 images take a long time. (resolution 1024x2048) at almost 100% CPU. I can't remember this was the case previously. It was much faster and also less CPU intensive.
2) How can I use docker with Nvidia GPU?
1) I installed everything again. (8 instances of Deepstack in Docker + settings in AItool) Same computer and same settings as previous installation. Somehow VMMEM consumes 90% of CPU of a Ryzen 5950X at 4.5Ghz. Even 5-10 images take a long time. (resolution 1024x2048) at almost 100% CPU. I can't remember this was the case previously. It was much faster and also less CPU intensive.
2) How can I use docker with Nvidia GPU?
Did you use this to create the containers?
docker pull deepquestai/deepstack:gpu
Do you really have 8 instances as in 8 ports for deepstack? What is the requirement for 8 deepstack instances may I ask?
whoami ™
Pulling my weight
I run 3 instances of native windows deepstack:gpu. In the later versions of AI tool there is an option to add more than one port which automatically runs separate instances of DS on those ports but it only works with native windows version of DS. Docker deepstack:gpu I couldn't tell you off the top of my head.
Did you have all 8 in docker before? I can imagine the VMMEM cpu being high with 8 containers running deepstack.
I have just moved my instances to windows deepstack to close down my docker desktop due to VMMEM cpu usage.
Just 2 docker instances also. cpu now sits at around 5-7% idle instead of 15% or so.
Docker did have more advantages though through portainer, cpu and memory settings, env settings for each instance etc.
Update: WSl2 has no constraints and will use anything available - you may want to throttle it back.
You can adjust this with a .wslconfig in the user folder.
Memory allocation to docker containers after moving to WSL 2 in Windows
I recently updated my Docker environment to run on WSL 2 on Windows. For setting memory allocation limits on containers in previous versions, I had option in Docker Desktop GUI under Settings->
stackoverflow.com
Last edited:
It's not just CPU load, but also slow processing, taking a few seconds per image for 1280x720 images. I have a 5950X16 core Ryzen CPU OC at 4.5ghz and 64GB memory. This is a really new and good computer. I can't imagine deepstack is that CPU intensive.
Yes, I had the 8 instances running before, and it resulted in much faster image recognition (it's a lot slower now for some reason).
Is Deepstack on docker still the best option, or can I better use the windows versions of CPU and GPU of deepstack?
Yes, I had the 8 instances running before, and it resulted in much faster image recognition (it's a lot slower now for some reason).
Is Deepstack on docker still the best option, or can I better use the windows versions of CPU and GPU of deepstack?
Last edited:
Did you have an intel before? I only have 4 cameras with low resolution on substreams.
I tried it out on a ryzen 5 3400g with integrated graphics. 8 cores. 32gm ram.
Muchh slower than my i7 7700 (8 virtual cores also). But your ryzen is far superior to my 5 3400g so I am surprised.
Moved my BI and deepstack back to the older intel. Intels are recommended over Ryzen.
I would try them on the windows deepstack just to compare.
Have you tried to limit the cores used on each deepstack instance? Command line or spin up portainer and just use the slider.
That helped mine on the ryzen. Well it helped the cpu from hitting max and bottlenecking. (At that time I had been using mainstreams on full HD and MODE high)