Search results
-
5.4.9 - June 25, 2021
Thought I saw that also, but glad someone else saw it too! I’m NOT nuts, well on this subject anyway. Sent from my iPhone using Tapatalk- kc8tmv
- Post #33
- Forum: What's New
-
5.4.9 - June 25, 2021
DeepStack: Server Error 100 all over the place with .3 release! Back to .2 quick! Sent from my iPhone using Tapatalk- kc8tmv
- Post #29
- Forum: What's New
-
Edge Vector vs Simple for motion
Kind of "bumping" this thread to see if in the latest versions ie. 5..4.9.x, edge vector is still the best in a non-zone crossing environment. Thoughts?- kc8tmv
- Post #5
- Forum: General BI Talk
-
5.4.9 - June 25, 2021
Yeah, I think I will turn off faces for now. Great idea, but agree very hit or miss and IMHO not very user friendly when it comes to management. Still like the idea of being able to add as many images of Bob in a "BOB" folder with any image name I want, and then when DS finds ANY hit on any...- kc8tmv
- Post #20
- Forum: What's New
-
5.4.9 - June 25, 2021
Were you trying the "Dark" custom model? I seem to always get the "Deepstack Server Error 100" when I try using "Dark" and Facial recognition at the same time.- kc8tmv
- Post #17
- Forum: What's New
-
AI push notifications
So set my framerate on both streams to 20 and on the substream set i Frame Interval to 20. Why is i Frame Interval grayed out on my main stream? -
AI push notifications
OK, now I am feeling more stupid than normal! I knew the substreams needed to be at lower resolution, but guessing I assumed to lower the frame rate as well. Below I have the mainstream configuration, substream configuration and then the full camera details from BI. Hope you don't mind give me... -
AI push notifications
I think I need a tutorial on this subject! Are these set in the camera configuration itself? -
DeepStack "Nothing Found"
So even after turning off facial recognition, I am still getting the "DeerStack: Server error 100" and "DeepStack has been restarted" error in the log. I guess I will remove the "dark.pt" file from the custom models folder / restart BI and see what happens now. -
DeepStack "Nothing Found"
I didn’t mind the update, just found the timing funny! Sent from my iPhone using Tapatalk -
DeepStack "Nothing Found"
Turned off facial recognition at 1:43 PM, so I will check later today if I am still getting DeepStack server errors. Coincidentally, after turning off facial recognition, I restarted BI and OF COURSE it picked up Ken's brand new 5.4.8.1 update. That seemed for fix the issue with closing out the... -
DeepStack "Nothing Found"
Are you using facial recognition and / or the custom "Dark" model? -
DeepStack "Nothing Found"
Seeing a lot of "DeepStack Sever Error 100" even after last nights update to 5.4.8.0. Also after running a clip through DS analysis with the new DeepStack Analyzer window open, I go to close the clip window and get the below error every time. Below are the log entries just for today. Note the... -
What is the maximum zoom I can get on a PTZ that looks similar the DH-SD29204T-GN ? I really do...
What is the maximum zoom I can get on a PTZ that looks similar the DH-SD29204T-GN ? I really do not like the looks of the SD49225XA-HNR. I am sure they are good, but I am willing to sacrifice a bit on the zoom to have the look and style of the DH-SD29204T-GN.- kc8tmv
- Profile post comment
-
I have a Dahua - DH-SD29204T-GN that I like, but would like to find something simialar but with...
I have a Dahua - DH-SD29204T-GN that I like, but would like to find something simialar but with about 12x optical zoom. Can this be found in this same form factor (not the big bulky white domes hanging from a stick!)?- kc8tmv
- Profile post
-
SMS alert message
I have done that, at least you got a response. I have never had a response. [emoji30] I am using node-red and passing it the fULL &JSON and I as m sure someone that knows JavaScript a LOT better and than I can do what I am looking for. If will post to a node-red forum and see if I can come up... -
SMS alert message
Love to have a macro option of just the "WINNING" object type (Max percentage). Don't care what model it came from, don't care about the percentage (already told it my thresholds) I just want to see "person", "People", "Dog", "Tank" (you never know!). I am a data guy, the more data granularity... -
Facial Recognition - Folders - Adding - Labels
1. Can / Should I have my "unknown" faces being deposited into the SAME folder that comes up when you click the "Faces" button on the Blue Iris Settings / AI Tab? To me it seems to make it easier to click there, find "unknown" (&CAM + timestamp) faces and then just click the edit icon and change... -
SMS alert message
Yep, so the "DeepStack" data is now BY DEFAULT (leave it blank)! I stopped being "lazy" and tried it also! :-O -
-
-
5.4.7 - May 25, 2021 - Support for DeepStack custom model files
That has been my experience as well. Not sure where there is a chair on my stairs, but yeah, it seems to tag anything it finds in analyzing. No complaints though. Sent from my iPhone using Tapatalk- kc8tmv
- Post #215
- Forum: What's New
-
5.4.7 - May 25, 2021 - Support for DeepStack custom model files
I know Ken doesn’t have release notes, but does anyone know what fixes are in 5.4.7.10? My prediction percentages seem to be higher. Anyone else notice this? Sent from my iPhone using Tapatalk- kc8tmv
- Post #196
- Forum: What's New
-
5.4.7 - May 25, 2021 - Support for DeepStack custom model files
I really think there is some conflict between the openlogos model and "faces" (facial recognition). I bet no one has both working at the same time. I think openlogos it the main issue.- kc8tmv
- Post #182
- Forum: What's New
-
5.4.7 - May 25, 2021 - Support for DeepStack custom model files
Maybe set + real-time images to 10 (or more) @ 1 second each?- kc8tmv
- Post #156
- Forum: What's New
-
Multiple "confidence" hits, which one wins?
Question on the confidence level. How does DeepStack / BI decide what to burn into the image and the final "confidence level"? Example below, "dark model" found People @ 75% and "objects" model found person @ 44%, yet only labeled image with "person:44%". I would have thought that the HIGHEST... -
5.4.7 - May 25, 2021 - Support for DeepStack custom model files
Yes, it normally does. Still lots of trial and error, but I see several of my own faces in there with a hat on, or different enough to not detect as me.- kc8tmv
- Post #128
- Forum: What's New
-
5.4.7 - May 25, 2021 - Support for DeepStack custom model files
My take on the {"api":"faces","found":{"success":true,"predictions":[{"confidence":0,"userid":"unknown", is that DS THINKS it is a face, but that you have not added THAT face to the "faces" DB (image set). Normally if mine does NOT find something it THINKS is a face, I get this...- kc8tmv
- Post #126
- Forum: What's New
-
Blue Iris Updates - Official Thread
Question on the confidence level. How does DeepStack / BI decide what to burn into the image and the final "confidence level"? Example below, "dark model" found People @ 75% and "objects model found person @ 44%, yet only labeled image with "person:44%". FYI, no idea where it found a chair going...- kc8tmv
- Post #371
- Forum: What's New
-
5.4.7 - May 25, 2021 - Support for DeepStack custom model files
The &JSON will send you something that looks like this: You can see WHICH models it is hitting. "objects" = build in DeepStack, "faces" = facial reccognition, "dark" = ExDark model...- kc8tmv
- Post #77
- Forum: What's New