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/

gwithers said:
I've been using 2.0-Beta since it was released over a month ago and this version has worked fine and without issue. So, there is nothing really wrong with that version from the object detection perspective. Maybe there are some other concerns (like the module install/uninstall functions don't seem to work) but nothing I use with BI. It is, after all, the docker container they have pushed as the one you would pull by default. If it was a problem, I assume they would have removed it weeks ago.
Click to expand...
How are you getting custom models working? I might just need a guide lol
 
harleyl7 said:
How are you getting custom models working? I might just need a guide lol
Click to expand...
Check out my post earlier in this thread (1,148). If a custom object detection model is not present in the explorer interface of CPAI. You would them need to add it as per my post. The normal ones from Mike should be there however but you need to change BI to look for what is showing in the CPAI explorer (I find it to be case sensitive). You also need that exact name, which again is case sensitive (ipcam-general.pt), to be in the folder you point BI to for your custom models. You also need to call that model (ipcam-general) in the per camera setup for AI if you only want to use that one model. Otherwise, it uses all of the custom models in the folder.
 
  • Like
Reactions: aesterling and harleyl7
VideoDad said:
@MikeLud1, I'm looking to retrain my delivery.pt model soon. It seems like the new version of CPAI has support for YOLOv5 6.2. Are you retraining your models for that? Also which size(s) are you targeting?
Click to expand...
I am going to retrain my models. using the current version of YOLOv5 7.0, the models do work with CPAI. I am going to do a nano, small, medium, and large version
 
  • Like
  • Love
Reactions: slidermike, looney2ns, tech101 and 4 others
joshwah said:
Any demo on how Super Resolution works? before and after?
Click to expand...

As far as I can tell, it is a utility that is included in CodeP explorer to enhance a picture or capture from one of your cams.

Here is screenshot 1 where you select the picture file.

Screenshot 2023-01-12 191534.png

This is after I clicked Super Resolution:

Screenshot 2023-01-12 185952.png

By right clicking then on the image you get a menu of options seen here:


Screenshot 2023-01-12 191905.png

That's about as far as I have gone with it :)
 
  • Like
Reactions: MikeLud1 and Vettester
I'm testing my GPU card so they enrolled me in the beta testing. Although the card has not been seen yet by CodeProject, the Engineers at CodeProject have been great on following up with me and getting more information about the card. By the way the results so far have been good with 2.0.4 beta.

I would like to test the Face Processing and the licensing Plate. I have an idea about the face Processing but with Plate recognition, anyone in Beta knows what settings do I need to use?

Thanks
 
joshwah said:
Any demo on how Super Resolution works? before and after?
Click to expand...

TBH I won't be expecting much, having played around with many different online AI based up scalers, image enlargers, super resolution etc tools, I’m yet to see anything that impresses me, they all seem to be about the same or worst as just zooming in with a decent image editing application (Photoshop/Gimp/ACDSee) and then applying some sharpening, smoothing – Most of the AI tools I have tried to date seem to apply either too much sharpening or smoothing leaving the image looking like a much worst but larger version of the original.
 
CrazyAsYou said:
TBH I won't be expecting much, having played around with many different online AI based up scalers, image enlargers, super resolution etc tools, I’m yet to see anything that impresses me, they all seem to be about the same or worst as just zooming in with a decent image editing application (Photoshop/Gimp/ACDSee) and then applying some sharpening, smoothing – Most of the AI tools I have tried to date seem to apply either too much sharpening or smoothing leaving the image looking like a much worst but larger version of the original.
Click to expand...

Same here, we all would love to have it work like on the FBI tv shows :rofl: BUT, as far as CPAI, it is working great for me. I now run 7 cameras on it out of 15 and the CPU version works great.
 
  • Like
Reactions: Vettester
Just upgraded from 2.0.2 to 2.0.4 and the service won't start. Before I installed the upgrade I quit BI and CPAI services. Installation said it was successful. The CPAI service path is correct. Any thoughts on what I need to try to resolve this? Thanks



 
Vettester said:
Can you start the service manually? You might want to try re-installing and use the repair option.
Click to expand...
Thanks for the reply... I did an unistall/reinstall and ended up with the same error. I'm trying the repair option and it's been on the step for 5 minutes, perhaps it's repairing something or it's hung.

Screen Shot 2023-01-15 at 3.48.29 PM.png
 
Repair just finished. Same error when I try and start the service manually. Status page shows that the server is not responding.

1673816761525.png
 
I've uninstalled CPAI and then restarted the PC. Oddly the CPAI service remains in the list of services. No idea why it's still there after an uninstall.
 
You must log in or register to reply here.
Top Bottom