PoseNet Demos

Contents

Demo 1: Camera

The camera demo shows how to estimate poses in real-time from a webcam video stream.

Demo 2: Coco Images

The coco images demo shows how to estimate poses in images. It also illustrates the differences between the single-person and multi-person pose detection algorithms.

Setup

cd into the demos folder:

cd posenet/demos

Install dependencies and prepare the build directory:

yarn

To watch files for changes, and launch a dev server:

yarn watch

If you are developing posenet locally, and want to test the changes in the demos

Cd into the posenet folder:

cd posenet

Install dependencies:

yarn

Publish posenet locally:

yarn build && yarn yalc publish

Cd into the demos and install dependencies:

cd demos
yarn

Link the local posenet to the demos:

yarn yalc link @tensorflow-models/posenet

Start the dev demo server:

yarn watch

To get future updates from the posenet source code:

# cd up into the posenet directory
cd ../
yarn build && yarn yalc push