Hi,
I am working on a video analytics solution using OpenVINO R3.2019, NCS2 on Intel i3 machine.
We use mutiple CCTV camera feeds and analyse the data, I encountered a scenario where I couldn't process mutiple Camera feeds(example - OpenVINO Face Recognition Solution for multiple cameras using NCS2).I
Is it true that, I can deploy only one model(example - One Face Recognition Model(face-reidentification-retail-0095)) inside NCS2 or Is there any possibility of renaming a model and adding to NCS2
Can you please let me know, if it's possible to process multiple video streams using one NCS2, as it's our business need.
Thanks
Guru
連結已複製
Hello, Guru.
In your scenario of multiple cameras utilizing please use models specifically designed for multiple input channels (e.g. multiple cameras or multiple video files).
https://docs.openvinotoolkit.org/2019_R1/_inference_engine_samples_multichannel_demo_README.html
Argument "-nc" specifies the number of processed cameras inputs.
Demos provided on the link above are only capable to deploy one model at a time, however, that's due to the way these demos are written, so please feel free to create your own sample in order to overcome this limitation.
As an example, single channel face detection demo provides up to 4 inference models.
Interactive Face Detection Demo - https://docs.openvinotoolkit.org/2018_R5/_samples_interactive_face_detection_demo_README.html
Thanks.
Best regards, Max.
Hi @chew__minjian ,
I have the same issue as yours.
Could you please tell me if you found a solution to that ?
I'll be greateful if you help me.
Thank you in advance.
regards.
