So how do we apply "custom" models to Senseai?

[aadje93]

Hi fellow BI users

I switched to Sense AI on my BI machine with a Nvidia P2000 and by default the timing seems slightly slower then the Mikelud "general" model for only person/car

Then i've read about the yolov5x-cls.pt model to be used for faster/higher accuracy determinations. But how do i apply this? How do i make codesense AI for BI use this specific model only?

i'm looking forward to your help

 

[sebastiantombs]

In the main console window for AI uncheck the box for "use default" model. In each camera AI page insert the model name in the box for model names. If you use more than one they are separated by commas, Spelling and case are critical for both model names and objects.

 

[wittaj]

In addition, this thread probably has answers to the questions you have/will have:

5.5.8 - June 13, 2022 - Code Project’s SenseAI Version 1 - See V2 here https://ipcamtalk.com/threads/codeproject-ai-version-2-0.68030/

NOTE: There is a separate thread from Code Project Version 2 here. https://ipcamtalk.com/threads/codeproject-ai-version-2-0.68030/ More formal support for Code Project’s SenseAI, now our preferred no-extra-cost AI provider over DeepStack. SenseAI is better supported by its developers and has...

ipcamtalk.com

 

[aadje93]

i think i have figured it out, noticing a longer analysis time using yolov5x-cls. But now how do i know what object types are in the models?

 

[Vettester]

i think i have figured it out, noticing a longer analysis time using yolov5x-cls. But now how do i know what object types are in the models?

5.5.8 - June 13, 2022 - Code Project’s SenseAI Version 1 - See V2 here https://ipcamtalk.com/threads/codeproject-ai-version-2-0.68030/

Been running CPAI for a few days and I'm not looking back to Deepstack. Everything is performing great and response times are better than Deepstack.

ipcamtalk.com