Human action recognition in deepstream

• Hardware Platform (Jetson / GPU)
• DeepStream Version
• JetPack Version (valid for Jetson only)
• TensorRT Version
TRT 8.0.1
• NVIDIA GPU Driver Version (valid for GPU only)

I’m working for human action recognition.

The final application will run on deepstream.
(1)This sample detection works only for one person in an image or multiple persons in an image.

(2)what changes are needed to make in deepstream to work for multiple persons in image

This is for one person. If you want to recognize multiple people, you can try to use a …->pgie(people detect)->sgie(action recognition)->… pipeline to do that.

Yeah that is the thing I’m planning to do. Thanks

Hello @edit_or Do you still need support for this topic? Or should we close it? Thanks.

Ok. I’ll ask again when I implement it.