thank you I was able to understand a lot but I already installed TensorFlow dependencies without creating a virtual env, should uninstall them and do exactly as u did? and if yes how can I do that?
Hey Ulas, I followed all your instructions in this video and downloaded everything successfully on my Raspberry PI 4, but I was wondering how do I get this to work with my RLC-520 Reolink IP camera, for my project I'm using object detection to notify me when people are detected. Thanks, Ricky
Thank you ! very clear guide. And it is working! it is not so easy to follow many steps, because there is always something different with versions, so it made me wonder - perhaps such a guide could be easier if all steps are delivered as a make file. A question if I may: the recognition is OK, but not so good. I was recognized as a "person" and the screen as "tv". Then I tried a toy train, and sometimes it was a "cat", sometimes "bed". the rate was 1.9-2 FPS. If only one thing is to be recognized, would the FPS rate be higher? Are you going to publish customer recognition video?
Hello congratulations on your tutorials How could it detect only one specific object (bottles)? and turn on a led if there is at least one bottle within the image
THANKS Ulas : i was able to follow the video and the github instructions and could run it succesfully. If i try to launch the ObjectDetectionPiCamera.py script with a mouse click i have some env problems ( ModuleNot Found Error for the cv2 module..... Do you think there is a simple way to add a few lines and make it running with a click ? .... I will figure out, in any case . But really thank you very much for the good guidance ! RD
You create a env that is why you can not run it outside of this env. There are two options. Run all commands again install cv2 without env. So you can run it with double click. Second you can create a bash script which brings up the env and run this python script for you, then you simply click this script. First one is more suitable. You welcome :)
Your output was choppy. A-frame drop was observed while you were demonstrating the output on the screen. How to get rid of this issue moreover does it possible to use a GPU on Raspberry Pi 4B ?
Thank you! Finally I have able to do it. I just have one question, after checking if tensorflow was installed I somehow got in the directory but did the rest outside the virtual environment would that be a problem?
ERROR: tensorflow-2.1.0-cp37-none-linux_armv7l.whl is not a supported wheel on this platform. This error shows up.. What to do??? My rasp pi is 4 model B with OS raspberry pi OS lite(Debian version: 11 (bullseye)) armv7l
thank you so much I successfully installed tensorflow in raspberry pi 4 B
great !!! But I think it is better to use tensorflow lite for Raspberry Pi 4
Hatasız çalışan bir video Teşekkürler Ulaş bey
thank you I was able to understand a lot but I already installed TensorFlow dependencies without creating a virtual env, should uninstall them and do exactly as u did? and if yes how can I do that?
Hey Ulas,
I followed all your instructions in this video and downloaded everything successfully on my Raspberry PI 4, but I was wondering how do I get this to work with my RLC-520 Reolink IP camera, for my project I'm using object detection to notify me when people are detected.
Thanks,
Ricky
Thanks for this video it helped me
Hello sir! Any idea how i can upload my own trained images with this?
Thank you ! very clear guide. And it is working! it is not so easy to follow many steps, because there is always something different with versions, so it made me wonder - perhaps such a guide could be easier if all steps are delivered as a make file. A question if I may: the recognition is OK, but not so good. I was recognized as a "person" and the screen as "tv". Then I tried a toy train, and sometimes it was a "cat", sometimes "bed". the rate was 1.9-2 FPS. If only one thing is to be recognized, would the FPS rate be higher? Are you going to publish customer recognition video?
Thanks for this video it helps me a lot! Did u make any video of how to training a new model, or how create a model with a different data? (23:06)
After I >>>import tensorflow it just returns :">>>" and I cannot do the next sudo. Any thoughts
ctrl+D
Hello
congratulations on your tutorials
How could it detect only one specific object (bottles)? and turn on a led if there is at least one bottle within the image
THANKS Ulas : i was able to follow the video and the github instructions and could run it succesfully.
If i try to launch the ObjectDetectionPiCamera.py script with a mouse click i have some env problems ( ModuleNot Found Error for the cv2 module.....
Do you think there is a simple way to add a few lines and make it running with a click ?
.... I will figure out, in any case .
But really thank you very much for the good guidance !
RD
You create a env that is why you can not run it outside of this env. There are two options. Run all commands again install cv2 without env. So you can run it with double click. Second you can create a bash script which brings up the env and run this python script for you, then you simply click this script. First one is more suitable. You welcome :)
Sir can we install on pi 3 this work
Sir,how to configure rpi to run .pb model and .pbtxt file ...should I follow the same steps as in the vdo?
Your output was choppy. A-frame drop was observed while you were demonstrating the output on the screen. How to get rid of this issue moreover does it possible to use a GPU on Raspberry Pi 4B ?
Thank you so much !!
Thank you! Finally I have able to do it. I just have one question, after checking if tensorflow was installed I somehow got in the directory but did the rest outside the virtual environment would that be a problem?
In the Tensorflow website it is recommended to use env. but it is up to you.
elinize sağlık. Subscribed!
Hi, may i know how to use a custom dataset? Thank you.
is this going to work using simple usb camera?
can tensorflow connect with speaker? it there any code for it to work?
buenas tardes, este tutorial solo sirve para rpi 4?
ERROR: tensorflow-2.1.0-cp37-none-linux_armv7l.whl is not a supported wheel on this platform. This error shows up.. What to do??? My rasp pi is 4 model B with OS raspberry pi OS lite(Debian version: 11 (bullseye)) armv7l
hi did you solve the issue ? I am encountering this same error
Hi how do I get it to only detect human only?
How to make it as speech output
Use tensorflow lite to get more fps