nvdsFaceObject for age and gender

while creating the deepstream pipeline for age and gender detection, how can use the NvDsFaceObject plugin (Deepstream Python API Reference β€” Deepstream Deepstream Version: 6.1.1 documentation) in the probe function. And how can I extract the age and gender from it?

First, you need to integrate your model into deepstream,
https://docs.nvidia.com/metropolis/deepstream/dev-guide/text/DS_using_custom_model.html
filling inference output parsed results into object meta data, then you can get it through adding probe function. about how to use probe function, refer to app deepstream-image-meta-test function: osd_sink_pad_buffer_probe

can you share the link to deepstream-image-meta-test

And instead of custom model I am trying to use facedetect by nvidia. The pretrained model for face detection

sources/apps/sample_apps/deepstream-image-meta-test/
as i know, facedetect model just detect face, not classify age and gender.

We working in Python. So can you please tell python related solution

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.
Thanks

It’s not c/c++ or python issue. if the model just detect face, do not have age or gender attributes, how can you get it?

1 Like