/Detector-mao

Primary LanguageJavaScript

Handpose demo

Contents

This demo shows how to use the handpose model to detect hands in a video stream.

Setup

cd into the demos folder:

cd handpose/demo

Install dependencies and prepare the build directory:

yarn

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

yarn watch

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

Cd into the handpose folder:

cd handpose

Install dependencies:

yarn

Publish handpose locally:

yarn build && yarn yalc publish

Cd into the demos and install dependencies:

cd demos
yarn

Link the local handpose to the demos:

yarn yalc link @tensorflow-models/handpose

Start the dev demo server:

yarn watch

To get future updates from the handpose source code:

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