Please provide complete information as applicable to your setup.
• Hardware Platform: Nvidia Jetson AGX ORIN • DeepStream Version: Deepstream 6.1.1 • JetPack Version : 5.0.2
**• TensorRT Version 8.4.1 **
**• Deepstream and all dependencies are running inside docker container built following this approach Link • Issue Type( questions, new requirements, bugs)
I am running a Deepstream pipeline with 2 RTSP input streams, nvinfer plugin for object detection, nvtracker (NvDCF) and RTSP sink.
We are noticing that certain tracking IDs are repeated in both camera streams even though they have different angles, and locating different target objects.
What could explain this behaviour?
Thank you for quick reply!
Primarily this fixes my issue programmatically.
With the display however I did notice few times repetitions of certain tracks IDs. When I print out the object_meta attributes I assert there is uniqueness per stream.
Could it be that whatever plugin responsable for mapping the ID to [0,1,…] is miscalculating the IDs?
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
Can you print the obj_meta->object_id? Share us the log if you see any miscalculation.