• Hardware Platform (Jetson / GPU) - > Jetson Orin AGX • DeepStream Version → 6.1 • JetPack Version (valid for Jetson only) → 35.1 • TensorRT Version → 8.4 • Issue Type( questions, new requirements, bugs)-> Question
I would like to run PeopleNet to detect people and also use the penultimate layer for feature extraction. I came across this post where the use case is to only do feature extraction so the penultimate layer is set as the output in the config.
I would like to do regular inference to detect people and use the penultimate layer for feature extraction from the bounding boxes found in the detection. How can I do this?
There is no update from you for a period, assuming this is not an issue anymore.
Hence we are closing this topic. If need further support, please open a new one.
Thanks
Hi @ctuchez1 ,
You can set penultimate layer as another output layer , then you can customize the post-processing to parser the output of the two output layers.