Skip to content

Handtrack.js Impact Notes

Jump to bottom
Victor Dibia edited this page May 13, 2019 · 4 revisions

This wiki page is used to keep track on relevant discussions on how handtrack.js is being used today by developers.

Overview

Handtrack.js was created as an accessible Library for Prototyping Handtracking Interactions Directly in the Browser. It enables developers who cannot or are not interested in building their own AI models for handtracking, to easily integrate this capability in their web applications. It achieves this, by providing a simple API which abstracts aways admin aspects of the model use process (model hosting, loading). Underneath, Handtrack.js is an object detection model (MobilenetV2, SSD) converted to @tensorflow js, wrapped in a js library with some helpful methods (load, detect, renderpredictions, start video, etc).

Some Stats

  • 1000+ stars on Github
  • Demo viewed over 10k times (as of 05/01.2019)
  • Blog Post viewed over 22k times. (as of 05/01.2019)
  • Over 100 code files on Github (Javascript, HTML) integrate the Handtrack.js library for prototyping.
  • Handtrack.js has been integrated in multiple student projects (including an in progress application to help children with learning disabilities point and interact with interfaces).

Discussion and Prototyping Ideas

When we think about it, some college students might come up with amazing things with hand track application. Might me something to do with serious projects counting on gesture password.

  • Other students have reached out personally indicating their use of Handtrack.js for prototyping school projects. Most of these students have mentioned limited AI and sometimes web experience.

Growth and Impact

Handtrack.js has been well received by the community. Some of these are referenced below.

  • Top trending Javascript Library Worldwide on Github.
  • Discussed and upvoted on Reddit
  • Posted on HackerNews (YCombinator)

Community Appreciation

News and Translations

Clone this wiki locally