Please provide complete information as applicable to your setup.
• Hardware Platform (Jetson / GPU) GPU
• DeepStream Version 6.2
• TensorRT Version 8.5.2
• NVIDIA GPU Driver Version (valid for GPU only) 530.30.02
• Issue Type( questions, new requirements, bugs)
In main config file, I have multiple sources:
[source0]
enable=0
...
[source1]
enable=1
...
[source2]
enable=1
...
source0 is not enabled, source1 and source2 are, but they get assigned source-id 0 and 1, respectively, instead of 1 and 2, as I would expect (as is in the source group name, for example [source1]).
I tried to utilize the camera-id
component, but it does not work at all.
I then tried to add some custom code in deepstream_source_bin.c
(function create_multi_source_bin()
):
if (useCameraIds)
{
bin->sub_bins[i].bin_id = bin->sub_bins[i].source_id = configs[i].camera_id;
configs[i].source_id = configs[i].camera_id;
}
else
bin->sub_bins[i].bin_id = bin->sub_bins[i].source_id = i;
And even though bin
and config
components now have source_id
set as camera-id
, when I get the NvDsFrameMeta.source_id variable in gstdsexample.cpp
, these changes somehow do not persist.
I assume I am missing the part of the code, where NvDsFrameMeta
gets its source_id
assigned (maybe it’s not open-source).
How would you recommend to utilize the camera-id
parameter, to be able to read its value in a buffer in gstdsexample.cpp
?