Add face recognition to deepstream pipeline after object detection

Hi,
I want to add face recognition model (tensorflow model) to deepstream pipeline after object detection element, in this case I want to replace carModel classification with face recognition, I want to give me output-tensor(tensor 128-d) from face-recognition model, Is it possible to do it with nvinfer plugin or I Have to use custom plugin ?

  1. Pls share your setup with us
  2. You can refer https://github.com/NVIDIA-AI-IOT/deepstream_reference_apps/tree/master/back-to-back-detectors