How to combine data according to sequence number of streams in deepstream? For example, the data dimension of video stream 1 is (256,704, 3), the data dimension of video stream 2 is (256,704, 3), and the data dimension of their combination is (2,256,704, 4). They are sorted in stream order and cannot be sorted randomly. Stream 1 precedes stream 2 in the newly arranged memory.
No. Seems your are not talking about vdeo streams but the tensor data for model input. Currently, if the tensor data comes from different video streams and you want combine them into the same batch in order, it is not supported by current nvstreammux. You may need to implement such function by yourself.
Do you have any examples? Custom function。thanks
I think you should be aware that the help documentation for deepstream refers to the data passing process.
https://docs.nvidia.com/metropolis/deepstream/dev-guide/index.html
If you need different channel data, you should be looking for the source_id in the metadata. if you need to differentiate between batch process streams, you should be looking at the bach_id. getting this id information will require you to do development parsing in your code.
There is no update from you for a period, assuming this is not an issue anymore. Hence we are closing this topic. If need further support, please open a new one. Thanks
Custom for what?
This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.