4.2 Running Pose Estimation on GPU

To run this example, perform the following steps:

Download the file mpf_an522_v2023v2_jb.zip from AN5522.
Copy the file linux_imx477_pose_estimation.py into the folder holoscan-sensor-bridge/examples
Open a new terminal and navigate to the holoscan-sensor-bridge folder by using the following command.
```
cd <PATH/TO/holoscan-sensor-bridge>
```
The next step involves downloading ffmpeg and ultralytics packages to run the pose estimation demo from Running Holoscan Sensor Bridge examples - NVIDIA Docs. Rather than going to the link above, type the following in the console
```
apt-get update && apt-get install -y ffmpeg
pip3 install ultralytics onnx
cd examples
yolo export model=yolov8n-pose.pt format=onnx
cd -
```
Note:
This conversion step only needs to be executed once; the yolov8n-pose.onnx file contains the converted model and is all that’s needed for the demo to run. The installed components will be forgotten when the container is exited; those do not need to be present in future runs of the demo.
To run the sensor bridge demonstration container, from a terminal in the GUI,

Note: Ignore the step if the docker is running already
```
xhost +
sh docker/demo.sh
```
To run the pose estimation demo,
```
python examples/linux_imx477_pose_estimation.py
```
Close the Holoviz application and exit the docker to terminate the application