[tool] [tutorial] Free AI Person Detection for Blue Iris

New release 1.71 10/27/20.

From the logs. I don't think its working.

DateFuncDetailLevelSourceAIServerCameraImageIdxDepthColorThreadIDFromFileFilename
10/28/2020 12:07:26 PMInsertHistoryItem StackDepth=14, TID=7, TCNT=28: v:\blueiris\images\dwsd.20201028_120725635.jpg: Object reference not set to an instance of an object. [NullReferenceException] Mod: InsertHistoryItem(History hist) Line:0:0ErrorAITOOLS.EXE192.168.0.179:5000DWSDdwsd.20201028_120725635.jpg25217TrueAITool.[2020-10-28].log
10/28/2020 12:07:30 PMOnCreated DebugAITOOLS.EXE192.168.0.179:5000DWSD253122TrueAITool.[2020-10-28].log
10/28/2020 12:07:30 PMOnCreated ====================== Adding new image to queue (Count=1): v:\BlueIris\Images\DWSD.20201028_120730637.jpgDebugAITOOLS.EXE192.168.0.179:5000DWSDDWSD.20201028_120730637.jpg254122TrueAITool.[2020-10-28].log
10/28/2020 12:07:30 PMImageQueueLoop Adding task for file 'DWSD.20201028_120730637.jpg' (Image QueueTime='0.0' mins, URL Queue wait='0ms', URLOrder=1 of 1, URLOriginalOrder=1) on URL ''DebugAITOOLS.EXE192.168.0.179:5000DWSDDWSD.20201028_120730637.jpg255025TrueAITool.[2020-10-28].log
10/28/2020 12:07:30 PMImageQueueLoop Done adding 1 total threads, ErrCnt=0, ImageProcessQueue.Count=0DebugAITOOLS.EXE192.168.0.179:5000DWSD256025TrueAITool.[2020-10-28].log
10/28/2020 12:07:30 PMDetectObjects Starting analysis of v:\BlueIris\Images\DWSD.20201028_120730637.jpg...DebugAITOOLS.EXE192.168.0.179:5000DWSDDWSD.20201028_120730637.jpg257120TrueAITool.[2020-10-28].log
10/28/2020 12:07:30 PMIsValidImage Image file is valid: DWSD.20201028_120730637.jpgDebugAITOOLS.EXE192.168.0.179:5000DWSD258223TrueAITool.[2020-10-28].log
10/28/2020 12:07:30 PMDetectObjects (1/6) Uploading a 218234 byte image to DeepQuestAI Server at DebugAITOOLS.EXE192.168.0.179:5000DWSDDWSD.20201028_120730637.jpg259123TrueAITool.[2020-10-28].log
10/28/2020 12:07:31 PMDetectObjects (2/6) Posted in 355ms, Received a 118 byte response.DebugAITOOLS.EXE192.168.0.179:5000DWSDDWSD.20201028_120730637.jpg260116TrueAITool.[2020-10-28].log
10/28/2020 12:07:31 PMDetectObjects (3/6) Processing results...DebugAITOOLS.EXE192.168.0.179:5000DWSDDWSD.20201028_120730637.jpg261116TrueAITool.[2020-10-28].log
10/28/2020 12:07:31 PMDetectObjects Detected objects:DebugAITOOLS.EXE192.168.0.179:5000DWSDDWSD.20201028_120730637.jpg262116TrueAITool.[2020-10-28].log
10/28/2020 12:07:31 PMOutsidemask ->Using found mask file C:\Users\BlueIris\Downloads\AITOOL-VORLONCD\cameras\DWSD.bmp...DebugAITOOLS.EXE192.168.0.179:5000DWSD263316TrueAITool.[2020-10-28].log
10/28/2020 12:07:31 PMOutsidemask ->ALL of the object is OUTSIDE of masked area(s). (9 of 9 points)DebugAITOOLS.EXE192.168.0.179:5000DWSD264316TrueAITool.[2020-10-28].log
10/28/2020 12:07:31 PMCreateDynamicMask * Starting new object mask processing *DebugAITOOLS.EXE192.168.0.179:5000DWSDDWSD.20201028_120730637.jpg265316TrueAITool.[2020-10-28].log
10/28/2020 12:07:31 PMCreateDynamicMask Current object detected: key=80102, name=Car, xmin=68, ymin=187, xmax=313, ymax=461, IsStatic=False, counter=0, camera=DWSD, create date: 10/28/2020 12:07:31 PM, imageName: DWSD.20201028_120730637.jpgDebugAITOOLS.EXE192.168.0.179:5000DWSDDWSD.20201028_120730637.jpg266316TrueAITool.[2020-10-28].log
10/28/2020 12:07:31 PMEquals Percentage Intersection of object: 98.56115% Current 'Car' key=83550, Tested 'Car' key=80102DebugAITOOLS.EXE192.168.0.179:5000DWSDDWSD.20201028_120730637.jpg267316TrueAITool.[2020-10-28].log
10/28/2020 12:07:31 PMCreateDynamicMask Found 'Car' (Key=80102) in masked_positions key=83550, name=Car, xmin=54, ymin=185, xmax=314, ymax=467, IsStatic=False, counter=1, camera=DWSD, create date: 10/27/2020 10:18:24 PM, imageName: DWSD.20201028_120730637.jpgDebugAITOOLS.EXE192.168.0.179:5000DWSDDWSD.20201028_120730637.jpg268316TrueAITool.[2020-10-28].log
10/28/2020 12:07:31 PMDetectObjects Result='DynamicMasked', Detail='Car (77.95%)', ObjType='Vehicle', DynMaskResult='Found', DynMaskType='Dynamic', ImgMaskResult='MajorityOutsideMask', ImgMaskType='Image'DebugAITOOLS.EXE192.168.0.179:5000DWSDDWSD.20201028_120730637.jpg2691DarkSlateGray16TrueAITool.[2020-10-28].log
10/28/2020 12:07:31 PMDetectObjects 1x masked, so it's an irrelevant alert.DebugAITOOLS.EXE192.168.0.179:5000DWSDDWSD.20201028_120730637.jpg270116TrueAITool.[2020-10-28].log
10/28/2020 12:07:31 PMDetectObjects (5/6) Performing CANCEL actions:DebugAITOOLS.EXE192.168.0.179:5000DWSDDWSD.20201028_120730637.jpg271116TrueAITool.[2020-10-28].log
10/28/2020 12:07:31 PMSave JSON Settings saved to REGISTRY and C:\Users\BlueIris\Downloads\AITOOL-VORLONCD\AITool.Settings.JSONDebugAITOOLS.EXE192.168.0.179:5000DWSD272216TrueAITool.[2020-10-28].log
10/28/2020 12:07:31 PMDetectObjects (6/6) Camera DWSD caused an irrelevant alert.DebugAITOOLS.EXE192.168.0.179:5000DWSDDWSD.20201028_120730637.jpg273116TrueAITool.[2020-10-28].log
10/28/2020 12:07:31 PMDetectObjects Total Time: 725ms (Count=8, Min=634ms, Max=2768ms, Avg=1183ms)DebugAITOOLS.EXE192.168.0.179:5000DWSDDWSD.20201028_120730637.jpg274116TrueAITool.[2020-10-28].log
10/28/2020 12:07:31 PMDetectObjects DeepStack (URL) Time: 355ms (Count=8, Min=316ms, Max=382ms, Avg=341ms)DebugAITOOLS.EXE192.168.0.179:5000DWSDDWSD.20201028_120730637.jpg275116TrueAITool.[2020-10-28].log
10/28/2020 12:07:31 PMDetectObjects File lock Time: 0ms (Count=1, Min=3ms, Max=3ms, Avg=3ms)DebugAITOOLS.EXE192.168.0.179:5000DWSDDWSD.20201028_120730637.jpg276116TrueAITool.[2020-10-28].log
10/28/2020 12:07:31 PMDetectObjects Image Queue Time: 343ms (Count=8, Min=265ms, Max=370ms, Avg=306ms)DebugAITOOLS.EXE192.168.0.179:5000DWSDDWSD.20201028_120730637.jpg277216TrueAITool.[2020-10-28].log
10/28/2020 12:07:31 PMDetectObjects Image Queue Depth: 1 (Count=8, Min=1, Max=1, Avg=1)DebugAITOOLS.EXE192.168.0.179:5000DWSDDWSD.20201028_120730637.jpg278216TrueAITool.[2020-10-28].log
10/28/2020 12:07:31 PMInsertHistoryItem StackDepth=14, TID=7, TCNT=28: v:\blueiris\images\dwsd.20201028_120730637.jpg: Object reference not set to an instance of an object. [NullReferenceException] Mod: InsertHistoryItem(History hist) Line:0:0ErrorAITOOLS.EXE192.168.0.179:5000DWSDdwsd.20201028_120730637.jpg27917TrueAITool.[2020-10-28].log
10/28/2020 12:07:35 PMOnCreated DebugAITOOLS.EXE192.168.0.179:5000DWSD280116TrueAITool.[2020-10-28].log
10/28/2020 12:07:35 PMOnCreated ====================== Adding new image to queue (Count=1): v:\BlueIris\Images\DWSD.20201028_120735637.jpgDebugAITOOLS.EXE192.168.0.179:5000DWSDDWSD.20201028_120735637.jpg281116TrueAITool.[2020-10-28].log
10/28/2020 12:07:35 PMImageQueueLoop Adding task for file 'DWSD.20201028_120735637.jpg' (Image QueueTime='0.0' mins, URL Queue wait='0ms', URLOrder=1 of 1, URLOriginalOrder=1) on URL ''DebugAITOOLS.EXE192.168.0.179:5000DWSDDWSD.20201028_120735637.jpg282025TrueAITool.[2020-10-28].log
10/28/2020 12:07:35 PMImageQueueLoop Done adding 1 total threads, ErrCnt=0, ImageProcessQueue.Count=0DebugAITOOLS.EXE192.168.0.179:5000DWSD283025TrueAITool.[2020-10-28].log
10/28/2020 12:07:35 PMDetectObjects Starting analysis of v:\BlueIris\Images\DWSD.20201028_120735637.jpg...DebugAITOOLS.EXE192.168.0.179:5000DWSDDWSD.20201028_120735637.jpg284121TrueAITool.[2020-10-28].log
10/28/2020 12:07:35 PMIsValidImage Image file is valid: DWSD.20201028_120735637.jpgDebugAITOOLS.EXE192.168.0.179:5000DWSD285225TrueAITool.[2020-10-28].log
10/28/2020 12:07:35 PMDetectObjects (1/6) Uploading a 217414 byte image to DeepQuestAI Server at DebugAITOOLS.EXE192.168.0.179:5000DWSDDWSD.20201028_120735637.jpg286125TrueAITool.[2020-10-28].log
10/28/2020 12:07:36 PMDetectObjects (2/6) Posted in 358ms, Received a 117 byte response.DebugAITOOLS.EXE192.168.0.179:5000DWSDDWSD.20201028_120735637.jpg287122TrueAITool.[2020-10-28].log
10/28/2020 12:07:36 PMDetectObjects (3/6) Processing results...DebugAITOOLS.EXE192.168.0.179:5000DWSDDWSD.20201028_120735637.jpg288122TrueAITool.[2020-10-28].log
10/28/2020 12:07:36 PMDetectObjects Detected objects:DebugAITOOLS.EXE192.168.0.179:5000DWSDDWSD.20201028_120735637.jpg289122TrueAITool.[2020-10-28].log
10/28/2020 12:07:36 PMOutsidemask ->Using found mask file C:\Users\BlueIris\Downloads\AITOOL-VORLONCD\cameras\DWSD.bmp...DebugAITOOLS.EXE192.168.0.179:5000DWSD290322TrueAITool.[2020-10-28].log
10/28/2020 12:07:36 PMOutsidemask ->ALL of the object is OUTSIDE of masked area(s). (9 of 9 points)DebugAITOOLS.EXE192.168.0.179:5000DWSD291322TrueAITool.[2020-10-28].log
10/28/2020 12:07:36 PMCreateDynamicMask * Starting new object mask processing *DebugAITOOLS.EXE192.168.0.179:5000DWSDDWSD.20201028_120735637.jpg292322TrueAITool.[2020-10-28].log
10/28/2020 12:07:36 PMCreateDynamicMask Current object detected: key=81208, name=Car, xmin=61, ymin=187, xmax=313, ymax=463, IsStatic=False, counter=0, camera=DWSD, create date: 10/28/2020 12:07:36 PM, imageName: DWSD.20201028_120735637.jpgDebugAITOOLS.EXE192.168.0.179:5000DWSDDWSD.20201028_120735637.jpg293322TrueAITool.[2020-10-28].log
10/28/2020 12:07:36 PMEquals Percentage Intersection of object: 98.92473% Current 'Car' key=83550, Tested 'Car' key=81208DebugAITOOLS.EXE192.168.0.179:5000DWSDDWSD.20201028_120735637.jpg294322TrueAITool.[2020-10-28].log
10/28/2020 12:07:36 PMCreateDynamicMask Found 'Car' (Key=81208) in masked_positions key=83550, name=Car, xmin=54, ymin=185, xmax=314, ymax=467, IsStatic=False, counter=1, camera=DWSD, create date: 10/27/2020 10:18:24 PM, imageName: DWSD.20201028_120735637.jpgDebugAITOOLS.EXE192.168.0.179:5000DWSDDWSD.20201028_120735637.jpg295322TrueAITool.[2020-10-28].log
10/28/2020 12:07:36 PMDetectObjects Result='DynamicMasked', Detail='Car (69.88%)', ObjType='Vehicle', DynMaskResult='Found', DynMaskType='Dynamic', ImgMaskResult='MajorityOutsideMask', ImgMaskType='Image'DebugAITOOLS.EXE192.168.0.179:5000DWSDDWSD.20201028_120735637.jpg2961DarkSlateGray22TrueAITool.[2020-10-28].log
10/28/2020 12:07:36 PMDetectObjects 1x masked, so it's an irrelevant alert.DebugAITOOLS.EXE192.168.0.179:5000DWSDDWSD.20201028_120735637.jpg297122TrueAITool.[2020-10-28].log
10/28/2020 12:07:36 PMDetectObjects (5/6) Performing CANCEL actions:DebugAITOOLS.EXE192.168.0.179:5000DWSDDWSD.20201028_120735637.jpg298122TrueAITool.[2020-10-28].log
10/28/2020 12:07:36 PMSave JSON Settings saved to REGISTRY and C:\Users\BlueIris\Downloads\AITOOL-VORLONCD\AITool.Settings.JSONDebugAITOOLS.EXE192.168.0.179:5000DWSD299222TrueAITool.[2020-10-28].log
10/28/2020 12:07:36 PMDetectObjects (6/6) Camera DWSD caused an irrelevant alert.DebugAITOOLS.EXE192.168.0.179:5000DWSDDWSD.20201028_120735637.jpg300122TrueAITool.[2020-10-28].log
10/28/2020 12:07:36 PMDetectObjects Total Time: 674ms (Count=9, Min=634ms, Max=2768ms, Avg=1126ms)DebugAITOOLS.EXE192.168.0.179:5000DWSDDWSD.20201028_120735637.jpg301122TrueAITool.[2020-10-28].log
10/28/2020 12:07:36 PMDetectObjects DeepStack (URL) Time: 358ms (Count=9, Min=316ms, Max=382ms, Avg=343ms)DebugAITOOLS.EXE192.168.0.179:5000DWSDDWSD.20201028_120735637.jpg302122TrueAITool.[2020-10-28].log
10/28/2020 12:07:36 PMDetectObjects File lock Time: 0ms (Count=1, Min=3ms, Max=3ms, Avg=3ms)DebugAITOOLS.EXE192.168.0.179:5000DWSDDWSD.20201028_120735637.jpg303222TrueAITool.[2020-10-28].log
10/28/2020 12:07:36 PMDetectObjects Image Queue Time: 292ms (Count=9, Min=265ms, Max=370ms, Avg=304ms)DebugAITOOLS.EXE192.168.0.179:5000DWSDDWSD.20201028_120735637.jpg304222TrueAITool.[2020-10-28].log
10/28/2020 12:07:36 PMDetectObjects Image Queue Depth: 1 (Count=9, Min=1, Max=1, Avg=1)DebugAITOOLS.EXE192.168.0.179:5000DWSDDWSD.20201028_120735637.jpg305222TrueAITool.[2020-10-28].log
10/28/2020 12:07:36 PMInsertHistoryItem StackDepth=14, TID=7, TCNT=25: v:\blueiris\images\dwsd.20201028_120735637.jpg: Object reference not set to an instance of an object. [NullReferenceException] Mod: InsertHistoryItem(History hist) Line:0:0ErrorAITOOLS.EXE192.168.0.179:5000DWSDdwsd.20201028_120735637.jpg30617TrueAITool.[2020-10-28].log
10/28/2020 12:12:06 PMCleanUpExpiredHistory Removing expired history: key=26539, name=Person, xmin=430, ymin=57, xmax=453, ymax=124, IsStatic=False, counter=0, camera=DWSD, create date: 10/28/2020 12:06:51 PM, imageName: DWSD.20201028_120650635.jpg which existed for 5.3 minutes. (max=5)DebugAITOOLS.EXE192.168.0.179:5000DWSD307112TrueAITool.[2020-10-28].log
10/28/2020 12:12:06 PMCleanUpExpiredHistory Removing expired history: key=30770, name=Person, xmin=380, ymin=79, xmax=390, ymax=108, IsStatic=False, counter=0, camera=DWSD, create date: 10/28/2020 12:06:46 PM, imageName: DWSD.20201028_120645635.jpg which existed for 5.3 minutes. (max=5)DebugAITOOLS.EXE192.168.0.179:5000DWSD308112TrueAITool.[2020-10-28].log
10/28/2020 12:12:06 PMCleanUpExpiredHistory Removing expired history: key=52452, name=Person, xmin=449, ymin=115, xmax=461, ymax=136, IsStatic=False, counter=0, camera=DWSD, create date: 10/28/2020 12:06:46 PM, imageName: DWSD.20201028_120645635.jpg which existed for 5.3 minutes. (max=5)DebugAITOOLS.EXE192.168.0.179:5000DWSD309112TrueAITool.[2020-10-28].log
10/28/2020 12:13:06 PMCleanUpExpiredHistory Removing expired history: key=38735, name=Car, xmin=446, ymin=82, xmax=489, ymax=120, IsStatic=False, counter=0, camera=DWSD, create date: 10/28/2020 12:07:21 PM, imageName: DWSD.20201028_120720570.jpg which existed for 5.8 minutes. (max=5)DebugAITOOLS.EXE192.168.0.179:5000DWSD310113TrueAITool.[2020-10-28].log
10/28/2020 2:29:41 PMReloadLogLoaded 408 lines in 9ms from C:\Users\BlueIris\Downloads\AITOOL-VORLONCD\AITool.[2020-10-28].log.InfoAITOOLS.EXE101TrueAITool.[2020-10-28].log
10/28/2020 2:29:43 PMStartPauseLogStopped auto log refreshInfoAITOOLS.EXE201TrueAITool.[2020-10-28].log
10/28/2020 2:29:43 PMchk_filterErrorsAll_Click Searching AITool.[2020-10-28].log...DebugAITOOLS.EXE301TrueAITool.[2020-10-28].log
10/28/2020 2:29:43 PMchk_filterErrorsAll_Click ...Found 17 of 410 lines that had an error for a total of 17 lines in AITool.[2020-10-28].log...DebugAITOOLS.EXE401TrueAITool.[2020-10-28].log
10/28/2020 2:29:43 PMchk_filterErrorsAll_ClickFound 17 errors in 0msInfoAITOOLS.EXE501TrueAITool.[2020-10-28].log
10/28/2020 2:29:45 PMReloadLogLoaded 413 lines in 8ms from C:\Users\BlueIris\Downloads\AITOOL-VORLONCD\AITool.[2020-10-28].log.InfoAITOOLS.EXE101TrueAITool.[2020-10-28].log
10/28/2020 2:29:45 PMStartPauseLogStopped auto log refreshInfoAITOOLS.EXE201TrueAITool.[2020-10-28].log
10/28/2020 2:29:45 PMchk_filterErrorsAll_Click Searching AITool.[2020-10-28].log...DebugAITOOLS.EXE301TrueAITool.[2020-10-28].log
10/28/2020 2:29:45 PMchk_filterErrorsAll_Click ...Found 17 of 416 lines that had an error for a total of 17 lines in AITool.[2020-10-28].log...DebugAITOOLS.EXE401TrueAITool.[2020-10-28].log
10/28/2020 2:29:45 PMchk_filterErrorsAll_ClickFound 17 errors in 0msInfoAITOOLS.EXE501TrueAITool.[2020-10-28].log
10/28/2020 2:29:46 PMReloadLogLoaded 418 lines in 14ms from C:\Users\BlueIris\Downloads\AITOOL-VORLONCD\AITool.[2020-10-28].log.InfoAITOOLS.EXE101FalseAITool.[2020-10-28].log
 
It's hard when the support is minimal. I gave up.
I think that you need to remember that both versions of this program have been done by people that do this as a hobby and not a job. I know this thread is very long but most of the answers are on here already.
 
Last edited:
  • Like
Reactions: aesterling
He hasn't published a new release for 27 days, but it looks like he's actively making changes. I wish I knew how to compile a github project, as he suggested a few pages back that somebody compile his work on their own if they want the latest rev.
I've given brief instructions on here how to do it. I didn't know how to, but actively searched the internet to find out how to do it. I've also warned that the latest comits are untested and that they may contain bugs.
 
Hey Guys,

is there any way of training the AI? It sometimes detects a Boat but for sure we dont even have water :) Also some Persons are detected but actually its a plant.

Most problems occur during night, especially when light is switched on due to motion detector (i mean a hardware device, not blue iris motion detection).


Besides that, is there some hardware which is known as doing fast AI processing? I am now using only 2 cameras, but it will be 5-7 later. At this moment processing one image (AI) takes 1,5-2 Seconds.
I am wondering how to operate lets say 5 cameras which might send images the same time - then it would take up to 10 seconds until all images (of one single motion detection) are processed?! But during this process, more images are sent to the folder...

try the masking in AI tools, will be the easiest solution (your plant is always there at night, so a mask will always ignore it)

yes, there is better hardware and software (using gpu for instance, or even a raspi works well), but whatever setup you have now, you should first check the following:
-make sure there are not multiple instances of deepstack running. have you shut down deepstack properly anytime your run the command? this will lag like crazy
-try to lower the camera resolution, or use a lower res stream for the AI. I've found 4 or 6MP no better than 2MP for all my cameras, and get massive storage and speed gains.
-run the deepstack mode on 'low'. im not sure how much worse this is, but it's good enough for most of us and usually quite a bit faster.
-how often are you triggering? even 1 image per second can be excessive on old hardware.
-are you running the latest deepstack version? seems to be big improvements in speed recently

and keep your eyes out for the gpu deepstack version updates if you have a gpu already. native windows deepstack version comes out soon
 
  • Like
Reactions: Saargebeat
So, I'm wondering if there is maybe some "circular logic" going on here, and that's why I get multiple texts?

First page instructions for single camera say:

2.3 Configure each camera
This step needs to done for every camera and every profile.
Go to Record, check 'JPEG snapshot each (mm:ss)', select the „aiinput“ folder from step 2.1, check the box 'Only when triggered' and set the interval to p.e. 0:04.0 (one image every 4 second).

Now go to 'Trigger', check 'Capture an alert list image' and set the Break time 'End trigger unless retriggered' to p.e. 3s, so that a short alert only causes one image to analyze. If you think that AI Tool might overlook an object "on first sight" because it's only party visible, you can also make the break time longer than the 4s interval. In this case, multiple images will be analyze by AI Tool.

So now, when motion is triggered, a snapshot is sent to the aiinput folder. AITool then grabs that image and sends it to deepstack. Say deepstack sees a person; it reports that to aitool. Attool then triggers the camera to trigger the camera.

Um...so then the camera is triggered, and when it's triggered, it goes thru the above process.

Seems like that could go on forever? I mean, it doesn't go on forever, so maybe I'm missing something? Can somebody please explain if this is correct, or if the first page post needs to be updated?
 
So, I'm wondering if there is maybe some "circular logic" going on here, and that's why I get multiple texts?

First page instructions for single camera say:



So now, when motion is triggered, a snapshot is sent to the aiinput folder. AITool then grabs that image and sends it to deepstack. Say deepstack sees a person; it reports that to aitool. Attool then triggers the camera to trigger the camera.

Um...so then the camera is triggered, and when it's triggered, it goes thru the above process.

Seems like that could go on forever? I mean, it doesn't go on forever, so maybe I'm missing something? Can somebody please explain if this is correct, or if the first page post needs to be updated?

I don't know whether multi clones / streams are required (but it makes the most sense to me that they are), but I think the bottom line is that-
any camera triggered by AItool (not by motion), should not be configured to save images into the aiinput folder, otherwise you have the situation you describe. I actually had this briefly at first, and it was looping and bogging down deepstack, doubling up on files and processing.

Now that I noticed your camera naming scheme - If you are attempting to follow 'the hook up' tutorial, he triggers the SD cam stream with mition, and triggers the HD cam stream with the AItool. these are effectively 'seperate cameras' in BI, and configured differently
 
Last edited:
I don't know whether multi clones / streams are required (but it makes the most sense to me that they are), but I think the bottom line is that-
any camera triggered by AItool (not by motion), should not be configured to save images into the aiinput folder, otherwise you have the situation you describe. I actually had this briefly at first, and it was looping and bogging down deepstack, doubling up on files and processing.
Yeah; I thought so too. So if that's the case, and we're both not just missing something, then that can only mean that the instructions for use with a single stream are simply written wrong. I guess if the people writing / working on aitool can't figure out how to explain their own software, then there's no way I'm gonna figure it out.
 
I have mine setup with clones. I find it much easier than single camera setup (and it doesn't use any additional resources in BI).
Works well like this (at least for me). I find single camera will generate too much junk in BI (alerts).
 
  • Like
Reactions: 105437 and cjowers
Yeah; I thought so too. So if that's the case, and we're both not just missing something, then that can only mean that the instructions for use with a single stream are simply written wrong. I guess if the people writing / working on aitool can't figure out how to explain their own software, then there's no way I'm gonna figure it out.
Stop bitching about folks who wrote and provided the ad on for free. They then spent time writing instructions that everyone has been able to follow but you. The developer(s) owe you nothing. Go figure it our yourself or pay for sentry or find another vms.
 
Not everybody has been able to follow the single camera instructions.
So figure out what you are doing wrong without being an ungrateful prick. Keep your passive aggressive crap to yourself and be thankful that folks shared their hard work with you for FREE! There are thousands of posts on this thread but you and another user count as "not everybody". Please.
 
  • Like
Reactions: seth-feinberg
I've been basically asking the same 2 questions for 2 months now, only because I don't know what I'm doing wrong (otherwise I wouldn't ask). Nobody has been able to answer them. How's that make me a prick?? Or ungrateful? If I didn't think this software was awesome, or right for me, i'd have moved on. I haven't, because I really want to know how to make it work.

All I want to know is: why do I get multiple texts with the same image (that doesn't show the subject of the detected motion), and how do I get a single text with a relevant image? That's all. Really.
 
I've been basically asking the same 2 questions for 2 months now, only because I don't know what I'm doing wrong (otherwise I wouldn't ask). Nobody has been able to answer them. How's that make me a prick?? Or ungrateful? If I didn't think this software was awesome, or right for me, i'd have moved on. I haven't, because I really want to know how to make it work.

All I want to know is: why do I get multiple texts with the same image (that doesn't show the subject of the detected motion), and how do I get a single text with a relevant image? That's all. Really.
This is the part that makes you a prick.
" I guess if the people writing / working on aitool can't figure out how to explain their own software, then there's no way I'm gonna figure it out. "

You are essentially asking how to setup the software. Do you want the ai tool developers to come by and install it for you? This is way over your head and skillset. Perhaps you can pay someone to do it for you.
 
Ah. Well, it seems you took that sentence differently than I intended it...it wasn't meant as an insult. I don't think I can probably explain it differently, so how about this instead:

I apologize.

I'll keep working with this trio of software (bi / aitool / deepstack) and hopefully get it working.
 
  • Like
Reactions: cjowers
Yeah; I thought so too. So if that's the case, and we're both not just missing something, then that can only mean that the instructions for use with a single stream are simply written wrong.

Could be, although i don't remember reading them. But this thread (in addition to the guide, tutorials, etc) was certainly helpful to get my setup working, hence why i'm now trying to help others who might be struggling. I guess if you get things working and can then suggest better wording for the guide, definitely do it.

People seem to be using these tools to accomplish slightly different things, so configurations and setups vary a lot. So there isn't just 1 method / 1 guide that will work for all. And it's hard to read the 112 pages of thread in 1 go.

But I would say this is the perfect place for questions and troubleshooting (most users aren't highly technical, otherwise we would all be programming our own tools, so I reckon your questions are more than fair). Just remember to be nice to the programming legends involved when you're frustrated.
 
I've been basically asking the same 2 questions for 2 months now, only because I don't know what I'm doing wrong (otherwise I wouldn't ask). Nobody has been able to answer them. How's that make me a prick?? Or ungrateful? If I didn't think this software was awesome, or right for me, i'd have moved on. I haven't, because I really want to know how to make it work.

All I want to know is: why do I get multiple texts with the same image (that doesn't show the subject of the detected motion), and how do I get a single text with a relevant image? That's all. Really.
As i have previously said you need to look in your AITool input folder and see what images are in there.
 
I've been basically asking the same 2 questions for 2 months now, only because I don't know what I'm doing wrong (otherwise I wouldn't ask). Nobody has been able to answer them. How's that make me a prick?? Or ungrateful? If I didn't think this software was awesome, or right for me, i'd have moved on. I haven't, because I really want to know how to make it work.

All I want to know is: why do I get multiple texts with the same image (that doesn't show the subject of the detected motion), and how do I get a single text with a relevant image? That's all. Really.
watch this new video it's very simple to understand