Jetson nano: camera stops working after a couple of inferences with yolov7

[farescharfii]

Hello,

I am running a yolov7 object detection model on jetson nano 2GB.
I am using a camera “Arducam imx219” pictures as input to the model.
The camera is working and I can get RGB arrays and stream the images.
The problem is that after feeding the input the yolov7 model for inference, the camera stops working after 10 iterations.
This is what i get during operation:

CONSUMER: Done Success
GST_ARGUS: Cleaning up
(Argus) Error Timeout: (propagating from src/rpc/socket/client/ClientSocketManager.cpp, function send(), line 137)
(Argus) Error Timeout: (propagating from src/rpc/socket/client/SocketClientDispatch.cpp, function dispatch(), line 91)
(Argus) Error Timeout: (propagating from src/rpc/socket/client/ClientSocketManager.cpp, function send(), line 137)
(Argus) Error Timeout: (propagating from src/rpc/socket/client/SocketClientDispatch.cpp, function dispatch(), line 91)
GST_ARGUS: Done Success
(Argus) Error InvalidState: Argus client is exiting with 2 outstanding client threads (in src/rpc/socket/client/ClientSocketManager.cpp, function recvThreadCore(), line 357)

Do you have any idea why the camera stops suddenly?
Thanks in advance for you help.
Fares

[DaneLLL]

Hi,
The model may be too heavy for Jetson Nano 2GB. We would suggest use Jetson Nano 4GB or later Jetson platforms such as Xavier, Orin.

[farescharfii]

Hello DaneLLL,

Thanks for the reply.
The model yolov7 is working on the jetson nano 2GB if the inputs are images or videos stored already in the jetson.

The problem appears only when i integrate the csi camera to get the pictures out of it.