How to complete a case study of dynamic input and output of deepstream

For example, crnn, some dynamic input models, now tensorrt should support, I want to find some cases

There is no update from you for a period, assuming this is not an issue any more.
Hence we are closing this topic. If need further support, please open a new one.

Can you elaborate what you want?

Hello, for example, I now put two models into the pipeline. I completed the entire process when I was doing OCR last year. I hope to perform face detection and recognition in the same way, but can the partial analysis of face recognition be a reference? . I use arcface