DOCSEARCHbyAlgolia
Submit Docs

jsfeat Documentation

repository·master·Indexed 25 days ago

https://github.com/inspirit/jsfeat

A JavaScript computer vision library for JS/HTML5 environments. It provides tools for image processing (grayscale, blurring, Canny edges), linear algebra (SVD, Eigen Vectors), multiview motion estimation (RANSAC, LMEDS), feature detection (Fast Corners, ORB), and object detection (HAAR, BBF).

Tokens
314
Snippets
0
Records
3
Agent score
33%

What's inside jsfeat

  1. Overview of jsfeat Computer Vision library

    master
    jsfeat is a JavaScript library designed for computer vision using modern algorithms in JS/HTML5 environments. It provides a suite of tools including image processing, linear algebra, multiview motion estimation, feature detection, and object detection.

  2. Explore jsfeat image processing and feature detection capabilities

    master

    jsfeat provides several modules for computer vision tasks:

    Image Processing

    • Grayscale conversion
    • Blurring (Box blur, Gaussian blur)
    • Histogram equalization
    • Edge detection (Canny edges)
    • Derivatives (Sobel, Scharr)

    Linear Algebra & Matrix Math

    • Matrix operations (transpose, multiply, etc.)
    • Solvers (LU/Gaussian elimination, Cholesky)
    • SVD decomposition, solver, and pseudo-inverse
    • Eigen Vectors and Values

    Multiview & Motion

    • Motion kernels (Affine2D, Homography2D)
    • Motion estimators (RANSAC, LMEDS)

    Feature Detection & Description

    • Fast Corners
    • YAPE and YAPE06 detectors
    • ORB feature descriptor

    Tracking & Detection

    • Lucas-Kanade optical flow
    • HAAR object detector
    • BBF object detector

  3. Install and import jsfeat

    master

    jsfeat can be used in both browser and CommonJS environments.

    • In the browser: The library attaches itself to the global window.jsfeat object.
    • In CommonJS/Node.js: The library is exported via module.exports, allowing you to use require('jsfeat').