Hello!
Is there any good example (C++ or Python) of using Detectnet_V2 model, trained in TAO, for inference with TensorRT?
I tried example from Jetson-Inference, but it doesn’t work with TAO Detectnet_V2 model. As I understood because DetectNet V1 and V2 has different “input shape”.
Kind regards.
Officially, you can find the python code in the tao open source link. See tao_tensorflow1_backend/nvidia_tao_tf1/cv/detectnet_v2/scripts/inference.py at main · NVIDIA/tao_tensorflow1_backend · GitHub.
You can also find the official inference github. See GitHub - NVIDIA-AI-IOT/deepstream_tao_apps: Sample apps to demonstrate how to deploy models trained with TAO on DeepStream. For example, Facedetect model is based on detectnet_v2 network. You can leverage it.
More info here. You can also login deepstream docker
docker run -it --rm nvcr.io/nvidia/deepstream:6.4-triton-multiarch /bin/bash
and find the postprocessing.
/opt/nvidia/deepstream/deepstream/sources/libs/nvdsinfer_customparser/nvdsinfer_custombboxparser.cpp
Also, there are some forum topics as reference.
How to run purpose built model Peoplenet on Jetson Nano in my own application?,
Incorrect Results When Using TensorRT Inference Server With TLT Model,
How to run Detectnet_v2_resnet18.trt without deepstream - #3 by Morganh.
Thank you!
This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.