Pose Estimation with DeepStream does not work

[sanzarak]

Hardware Platform : Jetson Nano
DeepStream Version: 5.0.1
JetPack Version : 4.4.1

I installed Pytorch and TorchVision as in the following guide.

PyTorch for Jetson

Below are pre-built PyTorch pip wheel installers for Jetson Nano, TX1/TX2, Xavier, and Orin with JetPack 4.2 and newer. Download one of the PyTorch binaries from below for your version of JetPack, and see the installation instructions to run on your Jetson. These pip wheels are built for ARM aarch64 architecture, so run these commands on your Jetson (not on a host PC). You can also use the pre-built l4t-pytorch and l4t-ml container images and Dockerfiles. PyTorch pip wheels JetPack 5

PyTo…

Pytorch: 1.7.0
TorchVision: 0.8.1

And I downloaded it from GitHub - NVIDIA-AI-IOT/deepstream_pose_estimation: This is a sample DeepStream application to demonstrate a human pose estimation pipeline. and compiled it.
Then I tried something like this:

sudo ./deepstream-pose-estimation-app …/streams/test.mp4 ./

The error message is as follows.

Now playing: …/streams/test.mp4
Opening in BLOCKING MODE
Opening in BLOCKING MODE
ERROR: Deserialize engine failed because file path: /home/dssdk/sources/apps/pose_estimation/pose_estimation.onnx_b1_gpu0_fp16.engine open error
0:00:06.342003619 18380 0x55b16b5000 WARN nvinfer gstnvinfer.cpp:616:gst_nvinfer_logger: NvDsInferContext[UID 1]: Warning from NvDsInferContextImpl::deserializeEngineAndBackend() <nvdsinfer_context_impl.cpp:1690> [UID = 1]: deserialize engine from file :/home/robopia/aiengine201/sources/apps/pose_estimation/pose_estimation.onnx_b1_gpu0_fp16.engine failed
0:00:06.342111590 18380 0x55b16b5000 WARN nvinfer gstnvinfer.cpp:616:gst_nvinfer_logger: NvDsInferContext[UID 1]: Warning from NvDsInferContextImpl::generateBackendContext() <nvdsinfer_context_impl.cpp:1797> [UID = 1]: deserialize backend context from engine from file :/home/robopia/aiengine201/sources/apps/pose_estimation/pose_estimation.onnx_b1_gpu0_fp16.engine failed, try rebuild
0:00:06.342146538 18380 0x55b16b5000 INFO nvinfer gstnvinfer.cpp:619:gst_nvinfer_logger: NvDsInferContext[UID 1]: Info from NvDsInferContextImpl::buildModel() <nvdsinfer_context_impl.cpp:1715> [UID = 1]: Trying to create engine from model files

Input filename: /home/dssdk/sources/apps/pose_estimation/pose_estimation.onnx
ONNX IR version: 0.0.4
Opset version: 7
Producer name: pytorch
Producer version: 1.3
Domain:
Model version: 0
Doc string:

WARNING: [TRT]: onnx2trt_utils.cpp:220: Your ONNX model has been generated with INT64 weights, while TensorRT does not natively support INT64. Attempting to cast down to INT32.

---

Someone please help
thanks.

And I can’t speak English well.

[AastaLLL]

Hi,

The error indicates that there is no TensorRT engine exist in your workspace.
And deepstream will convert one from the ONNX file directly.

If there is a file located at /home/dssdk/sources/apps/pose_estimation/pose_estimation.onnx*.
The pipeline should work normally.

Thanks.