Please provide complete information as applicable to your setup.
**• Hardware Platform (Jetson / GPU) Jetson
**• DeepStream Version 6.2
**• JetPack Version (valid for Jetson only) 5.0.2
**• TensorRT Version 8.4.1.5
**• CUDA 11.4.239
**• cuDNN 8.4.1.50
**• Issue Type( questions, new requirements, bugs) questions
**•Requirement details:I have a pipeline that includes
" nvurisrcbin+nvstreammux+nvnfer+nvtracker+fakesink"
questions1.Is there any example of data storage after nvstreammux that can provide reference? Does decoupling stored data have a significant impact on the latency of the entire pipeline?
questions2.Does the video frame data after nvurisrcbin in the above pipeline have a timestamp? If there is a timestamp, is it the camera exposure timestamp?
questions3.Are there any examples of models and new methods in the above pipeline that can be referenced in Nvidia?
What kind of data do you want to store? The original input video, the video with detected object bboxes and labels, or the images of the objects detected?
The video frame has its timestamp.
It depends on the camera driver but not the upper level framework. Normally it is the frame capture time.
There are lots of open source samples:
https://docs.nvidia.com/metropolis/deepstream/dev-guide/text/DS_C_Sample_Apps.html#id2
NVIDIA-AI-IOT/deepstream_reference_apps: Samples for TensorRT/Deepstream for Tesla & Jetson
NVIDIA-AI-IOT/deepstream_tao_apps: Sample apps to demonstrate how to deploy models trained with TAO on DeepStream
What do you mean by “new methods”? Methods for what?
What kind of data do you want to store? The original input video, the video with detected object bboxes and labels, or the images of the objects detected?
Save a batch of video frames processed by nvstreamux.
What do you mean by “new methods”? Methods for what?
Mainly updates to the same model.
Do you mean you want to store the batched frames in one file?
What does this mean?
Do you mean you want to store the batched frames in one file?
Copy a batch of data and use it for programs outside the pipeline.
How do you want the frame data to be stored? In seprarated files which contains the YUV or RGB data?