How to run Retinaface in DeepStream

• Hardware Platform (Jetson / GPU) GeForce GTX 1080
• DeepStream Version 5.0
• TensorRT Version
• NVIDIA GPU Driver Version (valid for GPU only) 440.33.01
• Issue Type( questions, new requirements, bugs) questions

I have run deepstream example code for “objectDetector_SSD” successfully.

But what I try to implement is about Retinaface run in deepstream.
I don’t kwnow how to edit the nvsparserbbox.cpp for Retinaface.
And is there more file I need to edit or complie?

My Retinaface model is trained by pytorch with mobilenet0.25.
Reference from :

Can some one give me some hints? please

I think you need to understand the postprocess of the model and then you can implement the postprocess parser in deepstream following /opt/nvidia/deepstream/deepstream-5.1/sources/libs/nvdsinfer_customparser

1 Like