Feature Request - Tensor Flow Coco SSD / Object Detection

[AndreasW79]

Hi there,

is there any plan to make object detection optinonal? Maybe its too much overhead for the cpu (in case of an rpi) but it would be great to recognize the vehicle type (i.e. car, bus, bicycle, people etc.).

Thanks in advance and thanks for your effort !

Best regards, andreas

[pageauc]

Only practical way is to write a separate program that could batch process the images. Doing it in real time would not be practical unless you had an add on hardware. Post processing would be possible to classify the saved images.

On Wed, Jan 26, 2022 at 5:09 AM Andreas W @.***> wrote:

Hi there,

is there any plan to make object detection optinonal? Maybe its too much overhead for the cpu (in case of an rpi) but it would be great to recognize the vehicle type (i.e. car, bus, bicycle, people etc.).

Thanks in advance and thanks for your effort !

Best regards, andreas

— Reply to this email directly, view it on GitHub https://github.com/pageauc/speed-camera/issues/109, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABNPKZEUJ2JLBCWKUKTWJH3UX7I7HANCNFSM5M2XUJCA . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.

You are receiving this because you are subscribed to this thread.Message ID: @.***>

-- YouTube Channel at https://www.youtube.com/user/pageaucp http://www.youtube.com/user/pageaucp GitHub Repository at https://github.com/pageauc

[pageauc]

BTW the ALPR (Automatic license plate reader that is included with speed camera is an example of post processing. It reads the database and processes images and updates the DB records as required. See code here https://github.com/pageauc/speed-camera/blob/master/alpr-speed.py

On Wed, Jan 26, 2022 at 8:31 AM Claude Pageau @.***> wrote:

Only practical way is to write a separate program that could batch process the images. Doing it in real time would not be practical unless you had an add on hardware. Post processing would be possible to classify the saved images.

On Wed, Jan 26, 2022 at 5:09 AM Andreas W @.***> wrote:

Hi there,

is there any plan to make object detection optinonal? Maybe its too much overhead for the cpu (in case of an rpi) but it would be great to recognize the vehicle type (i.e. car, bus, bicycle, people etc.).

Thanks in advance and thanks for your effort !

Best regards, andreas

— Reply to this email directly, view it on GitHub https://github.com/pageauc/speed-camera/issues/109, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABNPKZEUJ2JLBCWKUKTWJH3UX7I7HANCNFSM5M2XUJCA . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.

You are receiving this because you are subscribed to this thread.Message ID: @.***>

-- YouTube Channel at https://www.youtube.com/user/pageaucp http://www.youtube.com/user/pageaucp GitHub Repository at https://github.com/pageauc

-- YouTube Channel at https://www.youtube.com/user/pageaucp http://www.youtube.com/user/pageaucp GitHub Repository at https://github.com/pageauc

[AndreasW79]

Thanks thats a great hint and i will check it. BTW i added some quick and dirty implementation to push the csv data via mqtt, maybe this could find it´s way to an optional function? But i dont want to mix up things in this topic - sorry for that.

[johncblacker]

I'm beginning to think I should embrace the concept of recording video and post-processing it on a faster machine - say a p/c or faster SBC if you have such a device. Performance on the RPi is good but when you have to process several frames in succession it becomes a problem if you can't get the FPS up to a fairly high value; if your FPS value is too low then inaccuracy and missing very fast vehicles.