Please provide complete information as applicable to your setup.
• Hardware Platform (Jetson / GPU) GPU
• DeepStream Version 6.2
• JetPack Version (valid for Jetson only) -
• TensorRT Version 8.5.2
• NVIDIA GPU Driver Version (valid for GPU only) 525.105.17
• Issue Type( questions, new requirements, bugs) bug
• How to reproduce the issue ? (This is for bugs. Including which sample app is using, the configuration files content, the command line used and other details for reproducing)
I am using the Gst-nvvideo4linux2 inside my detection pipeline. But since the Update to DeepStream 6.2 the resulting video is broken. Here is a simplified example of the pipeline
filesrc location=input.mkv ! decodebin name=decode ! nvvideoconvert ! nvv4l2h264enc ! h264parse ! matroskamux ! filesink location=output.mkv
It seems the output video is not created properly, one reason might be wrong iframes. Checking output.mkv
with
ffprobe -v error -skip_frame nokey -show_entries frame=pkt_pts_time -select_streams v -of csv=p=0 test.mkv
results in a single iframe
0.000000
Two possible workarounds I have found are:
- using
x264enc
instead ofnvv4l2h264enc
. But this is much slower. - Setting the property
tuning-info-id=1
HighQualityPreset instead of the default2
LowLatencyPreset.
Both produce a correct video with ffprobe
output
0.000000
12.500000
25.000000
37.500000
50.000000
…
Am I missing something here when switching from DeepStream 6.1 to 6.2?