What does BI CPU usage look like on it's own? You mention CPU in total at 70...curious what just BI is.
[tool] [tutorial] Free AI Person Detection for Blue Iris
- Thread starter GentlePumpkin
- Start date
Make sure the original camera has clone master selected and then make sure that clone master is unchecked on the clones.
if the clones are working properly, they should have an * at the end of the camera name when viewing the cams in BI. No * means it is pulling two feeds instead of one.
if the clones are working properly, they should have an * at the end of the camera name when viewing the cams in BI. No * means it is pulling two feeds instead of one.
I don't set anything in for ADMIN-KEY or API-KEY.
When the service is running I go to and activate it that way.
And if you want to apply it in the commandline it would be -API-KEY=af-5667-yujhn-ssr
Sendt fra min SM-G973F med Tapatalk
Thanks, will try that.
If anyone else could oblige, curious what folks BI CPU usage is with all cameras and AI cameras running, then disable AI Cameras only, and what is that CPU Usage %? Trying to figure out if I have a setting off somewhere in my AI cameras causing them to use almost the same amount of CPU as the main cameras, even though BI isn't showing a bitrate usage for them.
If anyone else could oblige, curious what folks BI CPU usage is with all cameras and AI cameras running, then disable AI Cameras only, and what is that CPU Usage %? Trying to figure out if I have a setting off somewhere in my AI cameras causing them to use almost the same amount of CPU as the main cameras, even though BI isn't showing a bitrate usage for them.
Managed to get around the service issue (for some reason the Server service in services.msc was not running).
Getting back to this, are you guys using 8083 or 83 because that's the port you're using instead of 81?
Also, and weird, if I double click on Docker Desktop shortcut on my desktop....nothing happens? Presumably something should open?
Just wanted to share that I wrote something similar to AI Tool in PHP that natively runs on the Raspberry PI 3/4. On my install I have Deepstack and blueAI running on a PI 4. Processing an image takes about 650 - 700ms with the Intel stick on a USB 3 port. Image size is 740x640.
[GIN] 2020/07/17 - 09:23:53 | 200 | 700.99636ms | 192.168.1.8 | POST /v1/vision/detection
[GIN] 2020/07/17 - 09:23:56 | 200 | 676.721575ms | 192.168.1.8 | POST /v1/vision/detection
[GIN] 2020/07/17 - 09:23:59 | 200 | 691.511652ms | 192.168.1.8 | POST /v1/vision/detection
[GIN] 2020/07/17 - 09:24:02 | 200 | 673.042527ms | 192.168.1.8 | POST /v1/vision/detection
[GIN] 2020/07/17 - 09:30:02 | 200 | 642.752278ms | 192.168.1.8 | POST /v1/vision/detection
[GIN] 2020/07/17 - 09:39:27 | 200 | 706.303165ms | 192.168.1.8 | POST /v1/vision/detection
github.com
[GIN] 2020/07/17 - 09:23:53 | 200 | 700.99636ms | 192.168.1.8 | POST /v1/vision/detection
[GIN] 2020/07/17 - 09:23:56 | 200 | 676.721575ms | 192.168.1.8 | POST /v1/vision/detection
[GIN] 2020/07/17 - 09:23:59 | 200 | 691.511652ms | 192.168.1.8 | POST /v1/vision/detection
[GIN] 2020/07/17 - 09:24:02 | 200 | 673.042527ms | 192.168.1.8 | POST /v1/vision/detection
[GIN] 2020/07/17 - 09:30:02 | 200 | 642.752278ms | 192.168.1.8 | POST /v1/vision/detection
[GIN] 2020/07/17 - 09:39:27 | 200 | 706.303165ms | 192.168.1.8 | POST /v1/vision/detection
GitHub - bonifacio123/blueAI
Contribute to bonifacio123/blueAI development by creating an account on GitHub.
github.com
IAmATeaf
Known around here
The port you use is entirely up to you, I use 8083 as 8081 and 8082 are already being used by other software on my BI PC.
You can use any port provided it’s not already being used except ports such as 80, 443, 3389 plus a host of others that Windows uses.
You can use any port provided it’s not already being used except ports such as 80, 443, 3389 plus a host of others that Windows uses.
IAmATeaf
Known around here
Quick question. For the mask image I just free hand drew the masked area but reading another thread it looks like the mask is an array of rectangles.
So is what I have done to produce the mask image fine or do I need to redo and draw rectangles or is it AITools that makes use of the mask image and uses it as an overlay to then work out what can be ignored.
So is what I have done to produce the mask image fine or do I need to redo and draw rectangles or is it AITools that makes use of the mask image and uses it as an overlay to then work out what can be ignored.
danodemano
n3wb
- Joined
- Jul 31, 2017
- Messages
- 8
- Reaction score
- 1
I apologize if I missed it but it's really hard to search through this thread. I have everything set up and working perfectly, thank you for this awesome tool! I was wondering if there was an easy (programmatic) way to enable/disable the Telegram messaging functionality. A flaw I discovered today while my wife and I were sitting outside on the patio is that I was getting constant alerts as it detected a person. It wasn't wrong, but I would like a way to turn that off with a physical button. I have buttons I can pipe into Home Assistant/Node-RED I just need to know how to make changes to the settings and disable motion alerts. Thanks so much!!
Cameraguy
Known around here
- Joined
- Feb 15, 2017
- Messages
- 1,486
- Reaction score
- 1,134
Anyone have these memos working yet?
pmcross
Pulling my weight
pmcross
Pulling my weight
It does label it in BI, but only in the local console. Hopefully it gets added to UI3 and the mobile app.
Sent from my iPhone using Tapatalk