OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Developer Tools & SDKs

Handtrack.js

Handtrack.js is a JavaScript library for prototyping real-time hand detection in the browser, using a trained convolutional neural network to predict bounding boxes and return class names and confidence scores for image, video or canvas inputs.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Handtrack.js — OpenViST project profile in AI Developer Tools & SDKs.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Handtrack.js when exploring browser-based hand detection for a frontend prototype and the required output is a set of detected boxes, labels and confidence scores.

Who it is for

  • Web developers prototyping hand- or gesture-based interactions
  • Developers who want browser-side detection predictions from image, video or canvas elements

Documented uses

  • Request hand-detection bounding boxes from a browser media element
  • Use supported hand-pose labels as inputs for experimental web controls
  • Visualize detected hand locations in a browser application
Before you choose

Requirements & limits

The repository identifies the project as open source under the MIT license.

  • The documented target is browser frontend JavaScript; Node.js is not supported by the described current version.
  • The README describes 0.1.x as still in active development and warns that it may have issues.
  • The README says mobile browsers have not been extensively tested and notes known inconsistencies.
  • Version 0.0.13 detects hands only, while the newer 0.1.x version supports additional classes.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Load the library and request detections from an image

Set up the documented browser workflow to obtain detection predictions for an image element.

Before you start

  • A frontend JavaScript application running in a browser
  • An HTML image element to pass to the library
  • Load the library using a script tag or npm

Steps

  1. Import Handtrack.js using a script tag or install the handtrackjs npm package.

  2. Select the image element in the page.

  3. Call the asynchronous handTrack.load() method to obtain a model.

  4. Call model.detect(img) with the image element to receive predictions.

Expected outcome

An array of detections containing bounding boxes, class names and confidence scores.

Guide limitations

  • The documented library is for browser frontend JavaScript, not Node.js.
  • The README says the newer 0.1.x version may have issues; 0.0.13 is an older hands-only version.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
15 October 2026
Original website title
victordibia/handtrack.js · Official homepage title · retained in its source language
How we review sourcesSuggest a correction