How to delete some frames stored in GPU after decoding video?

I have some question when using deepstream,

  1. I can get the frames stored in GPU, like:
    batch_meta = pyds.gst_buffer_get_nvds_batch_meta(hash(gst_buffer))
    but how can I delete some frames I don’t want to process?

2.In deepstream, how to separate multiple input sources for subsequent processing independently?

For better understanding, can you help explain why you want to delete the frame? For the second question, running different deepstream program for different input source, is it an option?

Please also provide your setup and version information:
• Hardware Platform (Jetson / GPU)
• DeepStream Version
• JetPack Version (valid for Jetson only)
• TensorRT Version
• NVIDIA GPU Driver Version (valid for GPU only)

After preliminary analysis, we hope to delete some unnecessary frames and speed up the processing process. Does deepstream provide deletion operations in frames?
version information:
• GPU: 1050ti
• DeepStream Version:6.1
• TensorRT Version:8.2.5.1
• NVIDIA GPU Driver Version (valid for GPU only):510.47.03

  1. what is your media pipeline? which deepstream sample are you testing?
  2. how do you know some frames are unnecessary ?
  3. you can use Interval to dump some input frames, please refer to Gst-nvinfer — DeepStream 6.1 Release documentation

There is no update from you for a period, assuming this is not an issue any more.
Hence we are closing this topic. If need further support, please open a new one.
Thanks

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.