New CodeProject.AI Object Detection (YOLO11 .NET) Module

ipcamwalt said:
Does it work as a -v5 model with Mike's new module?
Click to expand...
I put a bluetooth motion sensor on my mailbox so I haven't used that model in awhile. For it to work with Mike's new model he would have to convert it. I don't use the full delivery model because there is to much traffic on my street.
 
Last edited:
Yeah that's what I thought. I hope he converts the delivery model, or can suggest a way to use a different module to use the delivery model. I have all the different delivery trucks coming up my driveway and it would be nice to know delivery vehicle vs. passenger.
 
You can run both at the same time and compare.

That being said, I believe v11 only detects people / vehicle and anything that doesn't look like that doesn't get identified.
 
Last edited:
  • Like
Reactions: Skinny1
MikeLud1 said:
I hope no news is good news. To all how is Object Detection (YOLO11 .NET) 1.4.0 module working for everyone?
Click to expand...
Nailed it! Vehicle and person detection is great. The ones I've got running -v5 give a bit more detail truck vs. vehicle, but v11 is more confident it's a vehicle than v5 is that it's a truck.

Great work Mike!
 
  • Like
Reactions: MikeLud1
MikeLud1 said:
I hope no news is good news. To all how is Object Detection (YOLO11 .NET) 1.4.0 module working for everyone?
Click to expand...
For me v1.4.0 is crashing and I don't have any object detection since I updated, just checked today and it crashed and it stays stopped after that.
Just swapped the appsettings, didn't change anything else, and the previous version was working.
Here is the exception message, in case it helps.
Bash: 
17:46:14:ObjectDetectionYOLO11Net.exe: CodeProject.AI.Modules.ObjectDetection.YOLO11.ObjectDetector[0]
17:46:14:ObjectDetectionYOLO11Net.exe:       ObjectDetection (.NET YOLO11) setting ExecutionProvider = "DirectML" with device 0
17:46:26:ObjectDetectionYOLO11Net.exe: Fatal error. 0xC0000005
17:46:26:ObjectDetectionYOLO11Net.exe:    at Microsoft.ML.OnnxRuntime.InferenceSession.Init(Byte[], Microsoft.ML.OnnxRuntime.SessionOptions, Microsoft.ML.OnnxRuntime.PrePackedWeightsContainer)
17:46:26:ObjectDetectionYOLO11Net.exe:    at Microsoft.ML.OnnxRuntime.InferenceSession..ctor(Byte[], Microsoft.ML.OnnxRuntime.SessionOptions)
17:46:26:System.__Canon, System.Private.CoreLib, Version=9.0.0.0, Culture=neutral, PublicKeyToken=7cec85d7bea7798e: ..ctor(System.String, Microsoft.ML.OnnxRuntime.SessionOptions)
17:46:26:ObjectDetectionYOLO11Net.exe:    at CodeProject.AI.Modules.ObjectDetection.YOLO11.ObjectDetector..ctor(System.String, Microsoft.Extensions.Logging.ILogger`1, Int32)
17:46:26:ObjectDetectionYOLO11Net.exe:    at CodeProject.AI.Modules.ObjectDetection.YOLO11.ObjectDetectionYOLO11ModuleRunner.GetDetector(System.String, Boolean)
17:46:26:ObjectDetectionYOLO11Net.exe:    at CodeProject.AI.Modules.ObjectDetection.YOLO11.ObjectDetectionYOLO11ModuleRunner.DoDetection(System.String, CodeProject.AI.SDK.Common.RequestFormFile, Single)
17:46:26:ObjectDetectionYOLO11Net.exe:    at CodeProject.AI.Modules.ObjectDetection.YOLO11.ObjectDetectionYOLO11ModuleRunner.Process(CodeProject.AI.SDK.Backend.BackendRequest)
17:46:26:ObjectDetectionYOLO11Net.exe:    at CodeProject.AI.SDK.Backend.ModuleRunnerBase.ProcessModuleCommands(CodeProject.AI.SDK.Backend.BackendRequest)
17:46:26:ObjectDetectionYOLO11Net.exe:    at CodeProject.AI.SDK.Backend.ModuleRunnerBase+d__57.MoveNext()
 
m4k0 said:
For me v1.4.0 is crashing and I don't have any object detection since I updated, just checked today and it crashed and it stays stopped after that.
Just swapped the appsettings, didn't change anything else, and the previous version was working.
Here is the exception message, in case it helps.
Bash: 
17:46:14:ObjectDetectionYOLO11Net.exe: CodeProject.AI.Modules.ObjectDetection.YOLO11.ObjectDetector[0]
17:46:14:ObjectDetectionYOLO11Net.exe:       ObjectDetection (.NET YOLO11) setting ExecutionProvider = "DirectML" with device 0
17:46:26:ObjectDetectionYOLO11Net.exe: Fatal error. 0xC0000005
17:46:26:ObjectDetectionYOLO11Net.exe:    at Microsoft.ML.OnnxRuntime.InferenceSession.Init(Byte[], Microsoft.ML.OnnxRuntime.SessionOptions, Microsoft.ML.OnnxRuntime.PrePackedWeightsContainer)
17:46:26:ObjectDetectionYOLO11Net.exe:    at Microsoft.ML.OnnxRuntime.InferenceSession..ctor(Byte[], Microsoft.ML.OnnxRuntime.SessionOptions)
17:46:26:System.__Canon, System.Private.CoreLib, Version=9.0.0.0, Culture=neutral, PublicKeyToken=7cec85d7bea7798e: ..ctor(System.String, Microsoft.ML.OnnxRuntime.SessionOptions)
17:46:26:ObjectDetectionYOLO11Net.exe:    at CodeProject.AI.Modules.ObjectDetection.YOLO11.ObjectDetector..ctor(System.String, Microsoft.Extensions.Logging.ILogger`1, Int32)
17:46:26:ObjectDetectionYOLO11Net.exe:    at CodeProject.AI.Modules.ObjectDetection.YOLO11.ObjectDetectionYOLO11ModuleRunner.GetDetector(System.String, Boolean)
17:46:26:ObjectDetectionYOLO11Net.exe:    at CodeProject.AI.Modules.ObjectDetection.YOLO11.ObjectDetectionYOLO11ModuleRunner.DoDetection(System.String, CodeProject.AI.SDK.Common.RequestFormFile, Single)
17:46:26:ObjectDetectionYOLO11Net.exe:    at CodeProject.AI.Modules.ObjectDetection.YOLO11.ObjectDetectionYOLO11ModuleRunner.Process(CodeProject.AI.SDK.Backend.BackendRequest)
17:46:26:ObjectDetectionYOLO11Net.exe:    at CodeProject.AI.SDK.Backend.ModuleRunnerBase.ProcessModuleCommands(CodeProject.AI.SDK.Backend.BackendRequest)
17:46:26:ObjectDetectionYOLO11Net.exe:    at CodeProject.AI.SDK.Backend.ModuleRunnerBase+d__57.MoveNext()
Click to expand...
Can you post your System Info, like the below

1762458370323.png
 
The model listing / naming is getting confusing .... I thought the only fully V11 one available was ipcam-general and if so why is there also a ipcam-general-v5 listed? Which should you use?
 
Spuds said:
The model listing / naming is getting confusing .... I thought the only fully V11 one available was ipcam-general and if so why is there also a ipcam-general-v5 listed? Which should you use?
Click to expand...
The ipcam-general model was trained using YOLO11 and the ipcam-general-v5 was trained using YOLOv5. All the models ending with -v5 are the same model that is in the Object Detection (YOLOv5 .NET) 1.14.0 module. I find the ipcam-general-v5 has a little be better accuracy

ipcam-general model

1762533588106.png

ipcam-general-v5
1762533705953.png
 
  • Like
Reactions: msantic and Skinny1
Hi everyone!
any chance to see a docker image integrating this? ;)
I'm running CPAI on one of my TrueNas Scale, so in docker mode from repository codeproject/ai-server:cuda12_2-2.9.7

And of course, thanks for your job, and your share with others!
 
zpn said:
Hi everyone!
any chance to see a docker image integrating this? ;)
I'm running CPAI on one of my TrueNas Scale, so in docker mode from repository codeproject/ai-server:cuda12_2-2.9.7

And of course, thanks for your job, and your share with others!
Click to expand...
See this post

MikeLud1

Post in thread 'New CodeProject.AI Object Detection (YOLO11 .NET) Module'

I just pushed a new version of the docker container. This version now has the Object Detection (YOLO11 .NET) 1.3.0 module. This version supports both YOLO11 and YOLOv5 models. All of my YOLOv5 models are included until I can train the YOLO11 versions on the models. The YOLOv5 models have a suffix of -v5, to use them you would for example ipcam-general is now ipcam-general-v5.

There are two tags for the container, they are the same container, you can use ether one.

1762044311622.png

1762044563574.png
  • Like
  • Love
 
  • Like
Reactions: looney2ns and zpn
Woah hang on here, where did CPAI 2.9.6 come from? Mine says it's at 2.9.5 and up-to-date.

Should I be updating it somewhere else?
 
So I've been having CPAI crashing a lot. The detections themselves are working.
That said, it's a new build/migration so "could" be something else but was using v11 on the new setup.
I just removed it and went to v5 native to see if that solves the issue maybe.

Maybe v11 doesn't like using the v5 detections if it get too busy or something? Nothing useful logged but figured would report it.
I've been working out other bugs on the machine too though so maybe unrelated and just appears that way. Guess we'll see how v5 does native...
 
You must log in or register to reply here.
Top Bottom