Preprocess for sgie

Accelerated Computing Intelligent Video Analytics DeepStream SDK
1

Please provide complete information as applicable to your setup.

• Hardware Platform (Jetson / GPU)–Xavier
• DeepStream Version– 5.0
• TensorRT Version --7.1

I am working on a pipeline which face detection as primary inference and generate face feature vector as secondary inference. Before the secondary inference, I have a face landmark detector which is added to user-meta-data. I want to align the faces using these landmarks points to feed the sgie. How do I pass the detected face crop to the sgie?

3

You can set the sgie as process-mode=2, so it will infer on object.
For the landmarks, I think you need to customized the nvinfer code.
You can refer DeepStream 5.0 nvinferserver how to use upstream tensor meta as a model input if you are using nvinferserver