- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi,
Can you please explain how I can run a model in GPU with camera in real time? I use Open VINO version : openvino_2019.3.379 for windows.
I want to connect camera and run model in real time.
Thanks and regards in advance,
Hanna
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi Hanna,
The classification_sample_async application only takes images as an input. We have other demos available that support a camera input. You can specify the camera input with “-i cam” when running the demo. Take a look at the following page for available demos and which models they support.
https://github.com/opencv/open_model_zoo/tree/master/demos
Regards,
Jesus
Link Copied
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi,
This is what I got with the command:
C:\Users\USER\Documents\Intel\OpenVINO\inference_engine_samples_build\intel64\Release>classification_sample_async -i "C:\Program Files (x86)\IntelSWTools\openvino\deployment_tools\demo\car.png" -m "C:\Users\USER\Documents\Intel\squeezenet1.1_FP16\squeezenet1.1.bat" -d GPU
[ INFO ] InferenceEngine:
API version ............ 2.1
Build .................. 32974
Description ....... API
[ INFO ] Parsing input parameters
[ INFO ] Parsing input parameters
[ INFO ] Files were added: 1
[ INFO ] C:\Program Files (x86)\IntelSWTools\openvino\deployment_tools\demo\car.png
[ INFO ] Creating Inference Engine
C:\Users\USER\Documents\Intel\OpenVINO\inference_engine_samples_build\intel64\Release>
This is the same result on CPU.
What can I do to run the models on GPU?
Thanks and regards in advance,
Hanna
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi Hanna,
The classification_sample_async application only takes images as an input. We have other demos available that support a camera input. You can specify the camera input with “-i cam” when running the demo. Take a look at the following page for available demos and which models they support.
https://github.com/opencv/open_model_zoo/tree/master/demos
Regards,
Jesus
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi Jesus.
About the comand:
Hi Hanna,
I just ran the following command and it worked in both CPU/GPU. Could you tell me which CPU you are using (example. Intel Core i5-6260U)?
classification_sample_async.exe -i "C:\Program Files (x86)\IntelSWTools\openvino\deployment_tools\demo\car.png" -m "C:\Users\jesus\Documents\Intel\OpenVINO\openvino_models\ir\public\squeezenet1.1\FP16\squeezenet1.1.xml" -d CPU
Could you also share the software version of the following components?
Python
cmake
Microsoft Visual Studio
Regards,
Jesus
I tried the command that you suggested and I got the same result as I mentioned in the command before.
My software : Python 3.6.5
Cmake 3.14.7
Microsoft Visual Studio 2019
My hardware: Intel Core i5-8265U
Thanks and regards in advance,
Hanna
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi Hanna,
Your last reply does not seem to be related to this thread. You have multiple threads going on, could you update your other thread?
Thanks in advance!
Regards,
Jesus
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi Jesus,
Runing model on GPU forum does not openning: page not found.
Thats why I answer you here.
Thanks and regards in advance,
Hanna
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi Abramov, Hanna,
That is interesting, I cannot open the other thread (Running model on GPU) as well as I receive the same "page not found" error. If possible please open a new thread with the same question you had in that thread so we can help you there, as these are two separate and different issues we don't want to get mixed. This way it's also easier to follow for everyone else looking at them and avoid confusion.
Regards,
Luis
- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page