Multi-pose estimation with TensorFlow

In this example, we will track the pose of a person during the video and draw keypoints over the image. In this case, we make use of the pipeless_ai_tf_models package to import the multipose estimation model, we don't even have to bring our own model, everything is ready to use.

Requirements

Pipeless: Check the installation guide.
Python OpenCV package. Install it by running:

pip install opencv-python

Run the example

Create an empty Pipeless project

pipeless init my-project --template empty # Using the empty template we avoid the interactive shell
cd my-project

Feel free to change my-project by any name you want.

Download the stage folder

wget -O - https://github.com/pipeless-ai/pipeless/archive/main.tar.gz | tar -xz --strip=2 "pipeless-main/examples/tf-pose"

Start Pipeless

The following command leaves Pipeless running on the current terminal

pipeless start --stages-dir .

Provide a stream

Open a new terminal and run:

pipeless add stream --input-uri "https://pipeless-public.s3.eu-west-3.amazonaws.com/woman-walking.mp4" --output-uri "screen" --frame-path "tf-pose"

Detecting Cats on a Video Object detection - YOLOv8 Python lib