MV3DTracker + ReID integration

Accelerated Computing Intelligent Video Analytics DeepStream SDK
1

I am using MV3DTracker with 2 input streams and I want to understand the behaviour of the ReID model.

is MV3DTracker use a ReID network to improve identity consistency. I enabled the ReID section in the tracker config and added the ReID model (as per the sample config). The tracker runs correctly, but I am not observing any visible difference in tracking or ID consistency after enabling ReID.

My questions:

  1. Does MV3DTracker actually use the ReID model for ID reassignment, or does it primarily use global 3D coordinates and trajectory smoothing for multi-stream association?

  2. If ReID is enabled, does MV3DTracker combine:

    • appearance similarity (ReID embedding),

    • global coordinates,

    • velocity/trajectory,
      for ID matching?
      Or is the ReID module optional and only used in specific pipeline configurations?

  3. I attached my tracker config file below.
    Could you please tell me if any parameters need to be changed for the ReID model to actually be used?

    config_tracker.txt (5.5 KB)

3

Here is the MV3DT sample: deepstream_reference_apps/deepstream-tracker-3d-multi-view at master · NVIDIA-AI-IOT/deepstream_reference_apps · GitHub

Here is the doc for MV3DT: Multi-View 3D Tracking (Developer Preview) — DeepStream documentation

The doc explain the details of target re-association and Tracklet Matching algorithm. Please check the doc.

4

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.