Closed akhil-faceintel closed 1 year ago
@akhil-faceintel sry didn't maintain this repo for a long time.
Thanks for explaining the states @zhouyuchong.
I have another question regarding user_meta information. Can you please explain how you add obj_user_meta_list information in obj_meta.
What does tensor_meta has to get the face embeddings?
@akhil-faceintel For Q1, I'm not adding obj_user_meta_list into obj_meta. The nvinfer will automatically add the raw tensor output into user_meta once you set output-tensor-meta=1
in config file.
For Q2, I'm not sure if I understand your question right. As I said above, once you get the raw output of a inference, you should do some postprocess and parse it like any other tensorRT inference.
Thanks for the explanation @zhouyuchong.
Great work!
Can you elaborate what does State means here?
@unique class FaceState(Enum): Init = 0 State1 = 1 State2 = 2 State3 = 3 State4 = 4