Using nvmultiurisrcbin with nvvidconv causes gst_nvvconv_transform: NvBufSurfTransform failed

Accelerated Computing Intelligent Video Analytics DeepStream SDK
,
1

Please provide complete information as applicable to your setup.

**• Hardware Platform: **Jetson
**• DeepStream Version: ** 6.3
• JetPack Version: 5.1.2
**• TensorRT Version: ** 8.5.2
**• NVIDIA GPU Driver Version: ** R35

Running pipelines of:
nvmultiurisrcbin gpu-id=0 width=2160 height=3840 ip-address=localhost batched-push-timeout=33333 max-batch-size=10 drop-pipeline-eos=1 live-source=1 uri-list=rtsp,rtsp ! nvvidconv ! video/x-raw, format=(string)BGRx ! fakesink

when have one RTSP on uri-list it work prefect

When I had two or more RTSPs, I received the error “gst_nvvconv_transform: NvBufSurfTransform Failed” and the process would stall.

Thanks

3

We can reproduce that on our side and are investigating it. Thanks

5

Could you try to use the pipeline below and check if that meets your needs?

... ! nvvideoconvert ! \
"video/x-raw(memory:NVMM),format=RGBA" ! \
nvvideoconvert ! "video/x-raw,format=RGBA"  ! ...
6

As a result of this change, I keep receiving the following errors:
/dvs/git/dirty/git-master_linux/nvutils/nvbufsurftransform/nvbufsurftransform_copy.cpp:571: => Batch sizes don’t match

/dvs/git/dirty/git-master_linux/nvutils/nvbufsurftransform/nvbufsurftransform.cpp:3586: => Copy Transformation Failed -1

7

You didn’t change your pipeline as I attached.

  1. You need to change the nvvidconv to the nvvideoconvert plugin because nvvidconv cannot support batched source.
  2. Use 2 caps filters as I attached.