Inference multiple models on a same camera/video source

I have run an object detection model and a lane detection model in different deepstream respectively, next step I want to run the object detection model and the lane detection model in one deepstream/gstreamer pipeline, I guess I should use two nvinfer after nvstreammux, but I don’t know how to deal the synchronous problem of two nvinfer/models(cause by the different framerate).
Is there any advice or solution for my problem.Thanks a lot.

Can you elaborate what kind of synchronous problem? Do you mean you set different “interval” values for the two gst-nvinfer elements?

Thanks for your reply, I have solved the problem by excuting the two models serially.

This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.