• Hardware Platform (Jetson / GPU)
Jetson Orin • DeepStream Version
6.1.1 • JetPack Version (valid for Jetson only)
5.0.2 • TensorRT Version
8.4.1-1+cuda11.4
Referring back to this. I need clarification
If I have a tracker and multiple different streams. When streams 1 and 2 start and both detect the first object, would the tracking ID of the first object be 0 and 1 correspondingly? Thus, Stream1-object0 and Stream2-obtect1
I am having that behavior in my Deepstream application and I am not sure if that is the correct behavior, I thought it would start a new tracker ID for each stream like Stream1-Object0 and Stream2-Object0
Another example: if I have 100 objects detected in Stream1 so: Stream1-Object0, Stream1-Object1,…, Stream1-Object99 and then in Stream2 I detect the first object, the tracker will assign an ID of 100: Stream2-Object100 instead of Stream2-Object0
@kesong upgraded to Deepstream 6.2, still the same. useUniqueID is present also in DS 6.1 and that creates a 64-bit long Unique ID when assigning tracker ID.
From the documentation:
Note that the incrementation of the lower 32-bit of the target ID is done across the whole video stream in the same NvMultiObjectTracker library instantiation. Thus, even if the unique ID generation is disabled, the tracker IDs will be unique for the same pipeline run. If the unique ID generation is disabled, and if there are three objects for Stream 1 and two objects for Stream 2, for example, the target IDs will be assigned from 0 to 4 (instead of 0 to 2 for Stream 1 and 0 to 1 for Stream 2) as long as the two streams are being processed by the same library instantiation.