ah you are correct. Just triggered it to try it out and it overwrites the text file with what is in the gui.
[tool] [tutorial] Free AI Person Detection for Blue Iris
- Thread starter GentlePumpkin
- Start date
- Sep 4, 2017
- 194
- 325
Deepstack can detect the following of which I chose the ones I considered most useful for AI Tool:
person, bicycle, car, motorcycle, airplane,
bus, train, truck, boat, traffic light, fire hydrant, stop_sign,
parking meter, bench, bird, cat, dog, horse, sheep, cow, elephant,
bear, zebra, giraffe, backpack, umbrella, handbag, tie, suitcase,
frisbee, skis, snowboard, sports ball, kite, baseball bat, baseball glove,
skateboard, surfboard, tennis racket, bottle, wine glass, cup, fork,
knife, spoon, bowl, banana, apple, sandwich, orange, broccoli, carrot,
hot dog, pizza, donot, cake, chair, couch, potted plant, bed, dining table,
toilet, tv, laptop, mouse, remote, keyboard, cell phone, microwave,
oven, toaster, sink, refrigerator, book, clock, vase, scissors, teddy bear,
hair dryer, toothbrush.
But I think deer is not among them.
person, bicycle, car, motorcycle, airplane,
bus, train, truck, boat, traffic light, fire hydrant, stop_sign,
parking meter, bench, bird, cat, dog, horse, sheep, cow, elephant,
bear, zebra, giraffe, backpack, umbrella, handbag, tie, suitcase,
frisbee, skis, snowboard, sports ball, kite, baseball bat, baseball glove,
skateboard, surfboard, tennis racket, bottle, wine glass, cup, fork,
knife, spoon, bowl, banana, apple, sandwich, orange, broccoli, carrot,
hot dog, pizza, donot, cake, chair, couch, potted plant, bed, dining table,
toilet, tv, laptop, mouse, remote, keyboard, cell phone, microwave,
oven, toaster, sink, refrigerator, book, clock, vase, scissors, teddy bear,
hair dryer, toothbrush.
But I think deer is not among them.
Whenever you may have time, would it be possible or difficult to have an option to include the object(s) and the confidence level into the telegram message?
Every now and then I start getting errors in processing with Deep Stack and I shut down both Deep Stack and AI Tool. This usually fixes things, but this time I'm not able to get it back up.
In the AI log tool folder, i'm seeing the following pattern of messages for all motion events:
[17.03.2020, 20:37:49.933]: Starting analysis of C:\BlueIris\DeepStack/UFlex2.20200317_203749907.jpg
[17.03.2020, 20:37:49.938]: (1/6) Uploading image to DeepQuestAI Server
[17.03.2020, 20:37:49.949]: (2/6) Waiting for results
[17.03.2020, 20:37:49.954]: (3/6) Processing results:
[17.03.2020, 20:37:49.961]: System.NullReferenceException | Object reference not set to an instance of an object. (code: -2147467261 )
[17.03.2020, 20:37:49.967]: ERROR: Processing the following image 'C:\BlueIris\DeepStack/UFlex2.20200317_203749907.jpg' failed. Failure in AI Tool processing the image.
Any thoughts on what to troubleshoot? I've tried starting each part in admin mode, clearing logs, reinstalling.
In the AI log tool folder, i'm seeing the following pattern of messages for all motion events:
[17.03.2020, 20:37:49.933]: Starting analysis of C:\BlueIris\DeepStack/UFlex2.20200317_203749907.jpg
[17.03.2020, 20:37:49.938]: (1/6) Uploading image to DeepQuestAI Server
[17.03.2020, 20:37:49.949]: (2/6) Waiting for results
[17.03.2020, 20:37:49.954]: (3/6) Processing results:
[17.03.2020, 20:37:49.961]: System.NullReferenceException | Object reference not set to an instance of an object. (code: -2147467261 )
[17.03.2020, 20:37:49.967]: ERROR: Processing the following image 'C:\BlueIris\DeepStack/UFlex2.20200317_203749907.jpg' failed. Failure in AI Tool processing the image.
Any thoughts on what to troubleshoot? I've tried starting each part in admin mode, clearing logs, reinstalling.
I gave up using this for a while due to issues with DeepStack. Had some time today so decided to give it another bash, looks like Deepstack on Windows is still an issue:
I can go to that IP address on a browser and reach the deepstack page, also tried localhost. Same thing unfortunately...
Running Windows 10 1909
Code:
[05.04.2020, 15:46:28]: ERROR: Processing the following image 'E:\BlueIris\AI_ID/DrwyAI.20200405_154448879.jpg' failed. Can't reach DeepQuestAI Server at 172.16.1.3:80.
[05.04.2020, 15:54:08]: ERROR: Processing the following image 'E:\BlueIris\AI_ID/DrwyAI.20200405_155227905.jpg' failed. Can't reach DeepQuestAI Server at 172.16.1.3:80.
[05.04.2020, 15:55:48]: ERROR: Processing the following image 'E:\BlueIris\AI_ID/DrwyAI.20200405_155232906.jpg' failed. Can't reach DeepQuestAI Server at 172.16.1.3:80.I can go to that IP address on a browser and reach the deepstack page, also tried localhost. Same thing unfortunately...
Running Windows 10 1909
Use a non-standard port like 82 or something. I have found using port 80 can cause weird conflicts since that the normal web server port.
Has anyone figured out definitively what low, medium, and high are referring to? Does high mean "high speed, lowest cpu utilization"? Or does high mean "highest level of analysis, slowest speed"? The website is super unclear.
Here's the text from the getting started website.
Performance
DeepStack offers three modes allowing you to tradeoff speed for performance. During startup, you can specify performance mode to be , “High” , “Medium” and “Low”
The default mode is “Medium”
Speed Modes are not available on the Raspberry PI Version
Here's the text from the getting started website.
Performance
DeepStack offers three modes allowing you to tradeoff speed for performance. During startup, you can specify performance mode to be , “High” , “Medium” and “Low”
The default mode is “Medium”
Speed Modes are not available on the Raspberry PI Version
@GentlePumpkin Thanks for making this, it is a great tool. I'm fairly new to any coding, but I would really like to take a look at the source and see if I can learn a few things and hopefully make some positive contributions to the project. I found the github link in previous pages of this thread, but it seems to no longer be there. Was is moved to a different location or removed for good?
- Sep 4, 2017
- 194
- 325
I'm very interested what these modes mean, too. But I have no clue aswell.
information for interested developers
Yes, I was annoyed by the time it took to maintain Github while the benefits for software quality weren't as expected. Additionally, I did not really like the MIT License I originally published it under, because it allowed p.e. make money with a closed-source software based on my and our collaborative efforts. That's my very personal stance, but I decided to select a license which as before allows everyone to freely modify the program, use it privately and so on, but requires source code for releases based on AI Tool. I published the repository again and it's still at the same location gentlepumpkin/bi-aidetection. It is available again and it will stay that way.
The general problem is that I have 10 things in mind that I would like to implement, but not even enough time to manage requests on Github within an adequate response time. This probably will change in about 1.5 years, but until then, please don't expect further development from my side. I currently intend to make no promises because in the end I may not have time to implement what was discussed. Being a newbie in programming, I tried to add as many source code annotations as possible (and helpful) and I hope they help everyone who discovers in the code. I will try to check Github every few weeks, review and published potential code submits, but that probably won't always be possible for me (same applies to the discussion here). If I don't react to your commit within a timespan you regard acceptable, please don't mind and just make your own fork. Later, it's possible to copy enhancements from froks to the main branch. It does really infuriate me that I'm unable to analyze and solve the problems some of us still experience with AI Tool and Deepstack.
Until then I'd like to thank everyone for your participation of all kind in shaping this very useful AI Tool. And I do intend to continue this way together with you
.By the way: In 1.5 years we probably will have an utterly different view on the current incredible global pandemic, hopefully a positive relief. Until then, please be care- and thoughful about what you do, maybe trust your gut feeling every now and then and I hope to see you again in no later than 1.5 years!
Your
GentlePumkpin
Last edited:
sansnil
n3wb
- Oct 7, 2016
- 12
- 3
I posted an issue to DeepStack support because I was getting false positives of objection recognition and one suggestion was to change from the low to med or high setting as it does a higher degree of analysis but it does take more CPU time so I'm inferring it means:
Low: lowest analysis, fastest speed
Med: medium analysis, medium speed
High: highest analysis, slowest speed
same here
[25.04.2020, 13:57:59]: ERROR: Can't write to cameras/history.csv!
[25.04.2020, 13:57:59]: ERROR: Can't write to cameras/history.csv!
[25.04.2020, 13:57:59]: ERROR: Can't write to cameras/history.csv!
[25.04.2020, 13:57:59]: ERROR: Can't write to cameras/history.csv!
This happen when blueiris do a massive delete on the folder used for AITool
why ???
I have the same issue, my DeepStack is running on another server but it just always says "can't reach"
This is in the log everything: System.Net.Http.HttpRequestException | An error occurred while sending the request. (code: -2146233079 )
Not sure what that means though, anyone have an idea what this could be?
Google says TLS connection but that seems unrelated.
pmcross
Pulling my weight
I’ve had this issue with running Deepstack on Windows; I believe that there was a bad update. I’ve had no trouble running it in Docker on Ubuntu though. Not sure which you’re running, but I can put a link to the version that has been working for me (on Windows) later today (I’m not by a computer right now).
Sent from my iPhone using Tapatalk
It's already running in a docker on a Linux server.
Are you running the latest Deepstack docker update?
pmcross
Pulling my weight
I have been primarily running on Windows, but I did some testing about two weeks ago on Docker on Ubuntu and had no problems. I was running the latest version of DeepStack and Docker at that time. Dumb question, but are you able to browse to the IP address and port using a browser to DeepStack?It's already running in a docker on a Linux server.
Are you running the latest Deepstack docker update?
Yes, that works just fine. Even tried disabling the firewall but that makes no difference.
I don't know how I can test Deepstack myself, do you know if there another tool I could us to test it?
pmcross
Pulling my weight
Not that I am aware of unfortunately. This problem has been plaguing a lot of users here. I will let you know if I am able to find a way to test.
ljw2k
Known around here
