Search results

There was a bug where if AI was on, saved alert images would no longer have the motion rectangles even if that option was selected if the AI found nothing/occupied. Upcoming minor version should fix that. This will help quickly identify at a glance the triggering motions that motivated the clip...

@bp2008 Any chance to have the AI memo be added to new alerts that are created after the UI3 window is open? It works on the "Confirmed alerts" view but the memos stay blank when on the regular "Alerts" view

Small addendum that I discovered: your Windows password can’t have special characters in it (at least a $) for Blue Iris to be able to use it. At least last I checked a few years ago. Maybe some escaping issue that Ken let slip through

CodeProject AI now supports Coral on Linux (Docker) x86 too. No need for Raspberry Pi. And you get less overhead and faster speeds than Windows and Raspberry Pi.

Timeline tab playback usually skips the first few seconds of every clip (which contains the actual important parts, making it useless). Non-Timeline playback is normal. Any ideas why?

Anyone else no longer seeing the "AI-verified" icons in the alerts list in UI3?

I noticed a bug that if an Alert happens while you're viewing UI3, the AI memo never gets populated for the newly added alerts on the list. It just stays blank. You have to refresh the UI3 page for the memos to show.

Nah if I don’t use HA, when these cams get triggered, the CPU usage flies to 80%+. With HA on (until they crash) it’s only a nice spike to 25% or so. Looks like Amcrest finally released a firmware update for these cameras, so I’ll give that a try and then hit support up to see what he can do...

Still getting these errors unless I turn hardware acceleration off for my Amcrest/Dahua cameras. The Reolinks work perfectly with Intel Beta. The Amcrests are choppy and unusable with Intel Beta, but they are smooth with plain Intel, but cause these errors and take down the whole system after...

I got the out of memory error today too, I’m on 5.4.7.11. Of course it was preceded by lots of CGadSocket errors and signals lost etc.

I wrote a script to copy the alert image elsewhere when certain AI objects are detected, like faces. Using the &ALERT_PATH macro (which is not even a path, it’s just a filename). It works most of the time, but BI craps out every now and then.

In some cases, I see that BI hits the DS endpoint faster than once a second, like 10 times a second, overwhelming my CPU. Other times it does the normal once a second thing. I haven't narrowed down when the former happens yet, or whether it's a good thing or not...

Just tried it out. It makes it so DeepStack doesn't start analyzing External triggers until it detects actual motion in the video. Useful for false positives in some cases, even if the static object detection should weed it out, but also helps so you don't run out of the "analyze additional...

5.4.7.11 adds a nice thing where rich image notifications for Apple devices are no longer letterboxed with black bars, so you see a full clear picture. +1 on unknown faces not being saved since about June 5.

Now that I consider all functionality as a whole, I think that the "Shield" functionality is the closest to this use case: either your system is "armed" or "disarmed", the natural binary choice inherent in alarm systems too. The Shield system lets you disable different parts when the system is...

So I only want push notifications when the house alarm is armed. I've been thinking about this. The only real conditions on push notifications seem to be: profile, input source, and AI objects. The obvious candidate is profiles. But I use profiles for different motion settings based on the time...

Were you seeing full usage before? How many cams did you have? Asking because I have 8 5MP cameras and I'm only seeing 2.6GB of RAM usage. Blue Iris itself, less than 1GB.

On 5.4.7.6 and the weird 2-second clips and clips ending before the trigger actually ended seems to be resolved.

One of the new 5.4.7.x releases added a feature where your app push notification attachment image sizes are set in a more sane manner, and also gives you the ability to switch the size between "Autosize", "960x480", and "480x480" in order to avoid the hated black bars on top of or along the...

Was informed by Ken that 5.4.7.7 will only check the Face endpoint if it detects a person, meaning DeepStack CPU usage should be cut down by up to half for most alerts. Interested in how he'll implement it (re-check the same image for a Face if it has a Person in it, or only check future images...

I’m still getting 3-second clips and clips randomly ending in the middle of triggers for no apparent reason with 5.4.7.3.

Any reason to use CBR instead of VBR?

Would this be the reason for weird 2-second clips and clips ending before their break time that I’m getting since a few versions ago? Or is that something else.

I noticed this yesterday independently. Motion will start in some leaves in the corner, and then unexpectedly, for a single frame or two, suddenly enlarge the object detection box to stretch into the middle of the screen for no apparent reason. Started happening recently but I wasn't sure if it...

Could be that the moment where a fast moving car was in view happens to fall in the 1 second interval between two DeepStack sends. The DS analyzer wouldn’t pick up on this coincidence because I don’t think it sends the exact frames it sent during the original incident. If that’s the case, I...

There is another bug as of 5.4.6.2 when creating Faces. Often, DeepStack won’t be able to find a face in your Face images. Even the ones that it spits out itself into your set Unknown Faces folder. But when you add a Face image and it fails to find a face in it and thus add it to the DeepStack...

Yeah, as of 5.4.6.2, even faces that are capitalized in the system ("Michael" instead of "michael") don't trigger alerts. Seems like a bug, reported it.

There's a bug where alerts don't recognize faces starting with capital letters. So "face1" and "face*" will work, but not "Face1" nor "Face*". Also, another bug where renaming faces to change capitalization doesn't actually register in the system as a change, so e.g. renaming "face1" to "Face1"...

Seems these "anamorphic" forced cameras may be cameras where the main stream and the sub stream have different aspect ratios, and Blue Iris forces the sub stream to change its aspect ratio to the main stream's ratio. (Except it's forced for some cameras where they are the same aspect ratio, so...

Great ideas. I wish there was a bug tracker of some sort, like a git repo except without the code. I know Ken's a bit old fashioned, so email's probably where he'll stay.