wonderful! i wonder if it was locked by some other program, maybe you opened in a text editor or something.
[tool] [tutorial] Free AI Person Detection for Blue Iris
- Thread starter GentlePumpkin
- Start date
I also have this need, I haven't been able to find any other way to accomplish this. Different Chat ID's would also enable custom notifications in the Telegram app i.e only with sound if it's really important. I have previously suggested it but if you add here this it will help: VorlonCD/bi-aidetection
blargrx
n3wb
Hello everyone, basic question before I get started on this project. Right now my family is using the BI ios app that pings notifications and previews when we have an alert. What does the end result of the project look like, you get alerts from telegram or from BI? My family would prefer to continue to get notifications only in BI app.
my understanding of how this all works is we are using 2 streams, SD is continuously recording and high sensitivity so it triggers to the jpg frequently to get analyzed by the deepstack AI. once confirmed, it triggers the HD recording for the alert. Would the BI app get notifications for both triggers on the SD and HD?
also what if we were not worried about our BI specs, plenty of HD and ram, can you just use the HD stream for both?
my understanding of how this all works is we are using 2 streams, SD is continuously recording and high sensitivity so it triggers to the jpg frequently to get analyzed by the deepstack AI. once confirmed, it triggers the HD recording for the alert. Would the BI app get notifications for both triggers on the SD and HD?
also what if we were not worried about our BI specs, plenty of HD and ram, can you just use the HD stream for both?
I would like to express my gratitude and admiration for the guys on this forum who have the interest and ability to make programs such as AI Tool by Gentle pumpkin and latterly Chris Dodge's Fork on this to Releases · VorlonCD/bi-aidetection. I have been a Blue Iris user for a number of years and it gets better and better. And now I am using VorlonCD/bi-aidetection with my Blue Iris.
My use for Blue Iris is not just survellance around my property for the usual stuff but mainly to see what wildlife is around during the night. In order to achieve the best results for that the motion trigger settings have to be set very sensitive with the inevitable numerous false triggers, especially when it rains. But combining judicious use of Cloned Cameras, VorlonCD Ai tool , DeepQuestAI and Docker things are getting better and better. I am sure I could improve things more but so far I am happy.
But I know from things I get notified about that Chris and his friends are making further inroads to improving VorlonCD adding things like amazon rekognition AI instead of DeepquestAI, thought to give better identification of items at night. But try as I might I cannot master the compiling of the various additions to VorlonCD AI releases and I have to rely totally on the Releases. Thanks again Chris and friends but please update the Release files with current addition when you can.
My use for Blue Iris is not just survellance around my property for the usual stuff but mainly to see what wildlife is around during the night. In order to achieve the best results for that the motion trigger settings have to be set very sensitive with the inevitable numerous false triggers, especially when it rains. But combining judicious use of Cloned Cameras, VorlonCD Ai tool , DeepQuestAI and Docker things are getting better and better. I am sure I could improve things more but so far I am happy.
But I know from things I get notified about that Chris and his friends are making further inroads to improving VorlonCD adding things like amazon rekognition AI instead of DeepquestAI, thought to give better identification of items at night. But try as I might I cannot master the compiling of the various additions to VorlonCD AI releases and I have to rely totally on the Releases. Thanks again Chris and friends but please update the Release files with current addition when you can.
Village Guy
Pulling my weight
Here you go, this is a very recent build that I am personally using and can vouch for VorlonCD AITool Version 2.0.195.7637 Build 30/11/2020
Last edited:
Thanks for that but as a matter of interest what is the version number, I am currently running Version 2.0.134.7620 Built on 11/11/2020
Ah Got it Version 2.0.195.7637 Build 30/11/2020. Installed now to see what changes there are. Thanks a lot.
Village Guy
Pulling my weight
I note from the last few posts that there is still confusion regarding Cloned cameras and Single \ Dual ( SD/HD setup)
I have made some notes below that I hope will help clear up the confusion and hopefully not make matters worse
Please feel free to contribute to the above text if you find it in error as I do not profess to be an expert on the subject.
I have made some notes below that I hope will help clear up the confusion and hopefully not make matters worse
- There is NO connection between Cloned camera setups and a setup for Single \ Dual stream. They are separate topics and NOT related.
- Cloned cameras ARE required if you only wish to record when your camera triggers on an AITool qualified event and do NOT want to record all events or continously.
- Single \ Dual Stream, BI when correctly configured has the ability to reduce CPU load by monitoring for movement in low resolution. When triggered the event JPEG is saved in maximum resolution to aid AITool in analysing the captured event image. This is especially useful as the CPU load is substantially reduced when you have a lot of cameras.
- Cloned Cameras The Cloned principle can be considered as a Master & Slave with the Master generating Triggers that are actioned by the Slave when AITool qualifies the JPEG generated by the Master. The Master and Slave are the same camera but setup within BI twice with one of them selected within BI as the Clone (copy of the master but setup slightly differently) By marking one configuration as the Clone, BI will not use additional resources over and above the Master camera. If neither is marked as a Clone, it will still work except BI will treat the camera as two separate cameras and CPU load will be increased accordingly. The Slave will always record in the selected resolution when AITOOL sends the command to the Slave camera.
- As a general rule if you intend to record every event you do NOT need to use Cloned cameras. IMHO it is also doubtful that Dual stream functionality brings much to the party as you will most probably be recording all events in HD and only use AITool to qualify events and index them.
Please feel free to contribute to the above text if you find it in error as I do not profess to be an expert on the subject.
Last edited:
How you set it up is really going to be based on what you want. You do not need to use 2 streams or you can, you don't have to use Telegram, you can use MQTT to push notifications, you can only use the AI on some or all your cameras etc. As much as you really do not want to ( assuming you were like me) I highly suggest you start at page 1 and read this whole thread, this project has really changed from Gentlepumpkins original and is actively being improved daily. There may be someone else that chimes in here and can help you more than I can but that is my advice.
Thanks @Tinbum I figured it out..well I think I figured it out, it took me numerous tries but when I cut and pasted everything into the folder it worked! I know it was just stubborn luck because if you asked me what I did I couldn't explain it.[tool] [tutorial] Free AI Person Detection for Blue Iris
Is there anything that can be done for this false positive? it happens a lot at night and can't be masked as it will render the entire camera useless because of the huge area. It is also so strange that the confidence level is so high. Try masking just the tree or the porch light and see if...ipcamtalk.com
So a little update- I "thought" all was good because I am getting alerts thru telegram (all cameras) then I go into AI-Tool to check the stats / history tabs and they are all blank and I have 300+ errors 
All the same 2 errors, No predictions... and Max queue.. Again I am getting alert images? ...Well crap.
Last edited:
Thanks, @Village Guy . You've been very helpful.
This is where I am still having some confusion. Specifically, I want to record the substream (SD) continuously and use it for alerts. I then only want to record the mainstream (4K) based on the substream alerts. Does your reference to "record continuously" mean just to continuously record the mainstream? Or can continuous recording be designated for the substream as well as using for alerts?
This is where I am still having some confusion. Specifically, I want to record the substream (SD) continuously and use it for alerts. I then only want to record the mainstream (4K) based on the substream alerts. Does your reference to "record continuously" mean just to continuously record the mainstream? Or can continuous recording be designated for the substream as well as using for alerts?
Village Guy
Pulling my weight
If you only want to record your HD stream when AITool qualifies the event then you need to configure your camera or cameras using the clone configuration. So you need two configurations setup for each of your cameras. One camera configuration will be used to monitor and trigger on an event. The other will receive commands from Aitool to start recording for the period that you have defined in the configuration.
Village Guy
Pulling my weight
Can I recommend that you first of all get your system functional before discussing the resolution of streams. It really makes no difference to the fundamental configuration and is confusing you. Once you have Aitool functional everything else will fall into place.
Need some help with compiling the code, Trying to learn but am having issues, can someone help shove me in the right direction- after reading the few posts here and a bunch of Googling I was able to use VS19 to get it compiled, I pulled the code from VorlonCD/bi-aidetection. once done I took a look at the settings.json file and I see a bunch of stuff is different from my current version. I Replaced everything with the new files (after making a B/U) It opened but none of my cameras or any other info came across, also there are some items that if I click on them the program errors out and closes, if were to copy the settings file from the previous version it would work but remove a bunch of items that are in the new (compiled version). So I am at a loss on what to do or how to proceed other than waiting for someone to eventually put a newer version out.
cscoppa
Getting the hang of it
- Dec 14, 2019
- 55
- 28
Pretty sure you're just supposed to replace setting.json with the new version. The actual settings for your cameras are in that _Settings folder. I just pasted in the build that
Village Guy complied, and it's working perfectly, didn't lose any of my settings. It was even smart enough to translate the .3 minutes I had for cooldown time to the new format of seconds.
Village Guy complied, and it's working perfectly, didn't lose any of my settings. It was even smart enough to translate the .3 minutes I had for cooldown time to the new format of seconds.
Last edited:
Last edited:
Your current situation sounds almost identical to mine. I'm running the VorlonCD fork v2.0.37.7608 10/30/2020. I would love to update, I keep checking the VorlonCD Github releases and it's still showing the version I'm running. As for wildlife, I live in a wooded area and there are several animals that come by almost nightly but several of them are not an option from Deepstack. I'd love to learn more about what @Chris Dodge is doing with Amazon Rekognition.
Village Guy
Pulling my weight
I'm guessing you simply downloaded and compiled a very old version. Make sure you are pointing to the version in development. If you are wanting a stable version, I would suggest you stay for the moment with the version I have posted. The reason Chris Dodge does not frequently post new versions might be that he does not wish to post code that may not be stable. FYI Just because you can download and compile the latest code does not mean you are running with the most stable and best version!That would be because @Village Guy knows what he is doing vs. me who does not
You are simply joining the Guinea Pig club
Last edited: