Using ActionRecognitionNet as a secondary gie

Please provide complete information as applicable to your setup.

• Hardware Platform (Jetson / GPU) Jetson
• DeepStream Version 6.0
I’m trying to configure a DeepStream pipeline using the normal deepstream-app reference. I want to use ActionRecognitionNet as the secondary gie, but without any pre-processing it doesn’t seem to work (I also tried adding the pre-processing but that didn’t work either). Since the pre-process group only supports primary gie’s is there any way to run this model without pre-processing or a way to connect the pre-processing to it?

We are developing the feature of pre-processing for SGIE.