Use external Re-Id feature vector in DeepSORT tracker

[MathiasDall]

Please provide complete information as applicable to your setup.

• Hardware Platform (Jetson / GPU)
GPU
• DeepStream Version
6.1
• JetPack Version (valid for Jetson only)
• TensorRT Version
• NVIDIA GPU Driver Version (valid for GPU only)
• Issue Type( questions, new requirements, bugs)
question, new requirements
• How to reproduce the issue ? (This is for bugs. Including which sample app is using, the configuration files content, the command line used and other details for reproducing)
• Requirement details( This is for new requirement. Including the module name-for which plugin or for which sample application, the function description)

Hi,

Would it be possible to add an option to attach external Re-Id feature vectors to an object’s metadata, and use those vectors when calculating similarity in DeepSORT INSTEAD of doing inference within the tracker module itself (such as mars-small128.pb etc.) ?

Ideally it should be possible to disable any model inference within the DeepSORT tracker entirely, and instead attach the Re-Id vectors to the input batch’s objects.

This would allow the user to run any Re-Id feature extractor as a SGIE (for instance, onnx models etc…) as well as providing a way of accessing the Re-Id vectors from outside the tracker (currently there is no way to access the vectors extracted in the DeepSORT tracker).

Thoughts…?

/M

[kesong]

Thanks your advice. I will discuss it internally and feedback here.

[cameron34]

Any updates on implementing this feature?

[kesong]

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

Export ReID vector from nvtracker to meta data is in our roadmap. Thanks.