Neelay Shah

@VijayaKumarKamma Can you provide more details on the system you are using and provide a gstreamer log by setting GST_DEBUG and /or provide a gdb back track? We do a...

Could you provide system details and a sample file? One thing to try to help isolate the issue would be to minimize elements in the pipeline (i.e. remove inference /...

I believe the issue is related to: https://github.com/dlstreamer/pipeline-server/issues/111 as it presents in the same way. We believe with the latest media and compute drivers that the issue is resolved. You...

@antoniomtz Is the goal to only produce bounding boxes with specified labels? Or to do secondary classification only on specified labels? yolo v3 will do inference on the entire frame...

metaggregate is used to aggregate multiple metadata onto the same frame from a single source (that is if you are doing detection + classification in different branches on the same...

DL Streamer supports batching accross pipelines using the model-instance-id parameter. If batching for inference is the main use case that should work (with some caveats on the individual streams having...

Couple thoughts: 1) you could add a `tee` element and multiple gvapython -> gvaclassify elements (you would also need to use metaggregate to put values back on the same frame....

The `video_frame` class has methods to retrieve the detections (region bounding boxes) and each region has an object id. You can find the class methods & descriptions here: https://github.com/dlstreamer/dlstreamer/tree/master/python/gstgva A...

@scalers-ai Independent of the ability to do remote processing of frame & results (which is a great feature add and something we've discussed as well) - can you give an...

@vtimofeev01 So we understand more clearly: are you asking for an example like gvaidentify (that is to add back the gvaidentify sample?) - Or are you asking for a sample...