Open-source alternatives to WebGazer

tracking.js is a browser computer vision library written in JavaScript for performing real-time image analysis and object tracking directly within a web browser. It functions as a real-time object tracker, a color tracking tool, and a face detection utility. The library enables the detection and monitoring of specific color ranges, human faces, and known visual patterns across consecutive video frames. It extracts visual features and descriptors from images to identify distinct landmarks for matching and tracking. The project covers broad computer vision capabilities, including the ability t

jeelizFaceFilter is a browser-based computer vision engine and WebGL face tracking library designed for AR filters and real-time facial movement tracking. It functions as a neural network face detector that identifies multiple faces and monitors mouth movements and rotation within a web browser. The system distinguishes itself through a model-swappable detection pipeline, allowing the exchange of neural network weights to balance accuracy and performance across different camera angles and devices. It features real-time lighting synchronization to match the illumination of 3D overlays with the

VDO.Ninja is a low-latency peer-to-peer media routing service and video streaming platform designed to integrate remote audio and video feeds into professional production workflows. It functions as a WebRTC broadcast integration tool and studio controller, allowing for the direct transmission of high-definition media between publishers and viewers with minimal delay. The platform distinguishes itself through extensive protocol bridging, converting between WebRTC, WHIP, WHEP, SRT, and RTMP to ensure compatibility across diverse network environments and professional studio software. It includes

ArtPlayer is a modular, browser-based media library designed for embedding and controlling video content within web applications. It functions as a programmable web component that provides developers with granular control over the playback engine, event handling, and the underlying video elements. The player distinguishes itself through a plugin-based architecture that allows for the injection of custom functionality, such as specialized visual effects, advertising, or quality controls, without modifying the core codebase. It utilizes a layered overlay system for interactive UI design, enabli

Screenity is a browser-based screen recorder designed to capture screen activity and audio directly within a web browser. It functions as a privacy-focused capture tool that handles data locally and includes a web-based video editor for basic media refinement. The project distinguishes itself through real-time screen annotation tools, allowing users to draw shapes and arrows or zoom into specific areas during a recording. It also provides specialized privacy controls to blur sensitive information and apply backgrounds to camera feeds. The tool covers a broad range of media capabilities, incl

OpenFace is an affective computing framework and facial behavior analysis toolkit designed for the extraction of facial features and the recognition of muscle movements to analyze human emotional behavior. It provides a research platform for analyzing human emotion and social interaction patterns through computer vision. The software implements a suite of tools for detecting facial landmarks, calculating head pose in 3D space relative to a camera, and tracking eye gaze by analyzing eye position and orientation. It also includes capabilities for facial action unit recognition to identify speci

vrtk is a virtual reality interaction framework and locomotion system designed to build consistent 3D object interactions and user interfaces across different hardware. It provides a hardware-agnostic movement model that decouples player navigation logic from specific controller inputs to ensure cross-platform compatibility. The project includes a spatial UI kit for rendering 2D and 3D interactive menus and physical controls specifically for head-mounted displays. It utilizes an event-driven input layer to translate physical pointer and touch data into virtual object events and actions. The

VRTK is a virtual reality interaction framework and UI toolkit. It provides a collection of components for integrating 2D and 3D interface elements, input handling systems, and navigation frameworks into virtual environments. The project includes a physics engine integration layer for managing body collisions and the behavior of interactive objects. It also provides a locomotion system specifically designed to implement movement techniques that reduce motion sickness in virtual spaces. The framework covers 3D object interaction, virtual UI design, and the development of user-driven events. T

jsmpeg is a JavaScript MPEG1 video decoder and canvas-based video renderer. It functions as a client-side media decoder that processes video frames in JavaScript, rendering them to an HTML5 canvas without relying on native browser video elements. The project provides a low-latency video player capable of receiving binary media data over WebSockets for real-time streaming. It also supports the progressive loading of static video files via Ajax to allow playback to begin before a file is fully downloaded. The system includes capabilities for playback state and event management, allowing for th

This project is a JavaScript library designed for real-time face detection directly within a web browser. It functions as a machine learning model wrapper that enables developers to identify and track human faces in live video streams without the need for backend server processing. The library utilizes browser-native media access to stream raw camera data into application memory, where it performs pixel-level analysis. By leveraging a tensor-based inference engine and web-assembly acceleration, the tool executes complex neural network calculations locally to achieve high-performance computer

This is a WebRTC video calling platform and browser-based video conferencing tool. It provides an instant communication room generator and an embeddable video call widget for hosting group audio and video calls directly in the browser. The platform enables a no-authentication communication workflow, allowing for the creation of unique, shareable links to establish private video sessions without requiring user account registration or software installation. The system supports real-time group video calling and provides capabilities for integrating the communication interface into external web

Real-Time-Person-Removal is a web-based computer vision application designed to identify and remove human figures from live video streams. Using TensorFlow.js, the tool functions as a real-time background subtraction system that analyzes scene composition to isolate static backgrounds from moving people. The project enables browser-based computer vision by processing webcam video feeds directly in the client. It utilizes machine learning to differentiate between dynamic scene elements and the background, allowing for the real-time removal of people from the visual field.

Annyang is a speech recognition library and web speech API wrapper that enables the integration of voice command interfaces into websites. It functions as a browser-based voice controller, mapping spoken phrases and regular expressions to specific JavaScript functions to trigger application actions. The library provides mechanisms for voice command mapping and simulation, allowing developers to associate spoken text with executable callbacks. It includes tools for command variable extraction using regular expression capture groups, which allows specific words from a spoken phrase to be passed

Mirotalk is a self-hosted video conferencing platform and peer-to-peer communication server. It utilizes WebRTC for high-resolution video calls, screen sharing, and real-time audio streaming, operating as a decentralized system that routes media streams directly between participants to ensure privacy and low latency. The platform integrates conversational language models and speech recognition to provide real-time transcription and automated meeting tasks. It also functions as a white-label collaboration suite, allowing users to apply corporate visual identities across the interface and sourc

This project is a collection of pre-trained machine learning models and conversion pipelines designed for running inference directly in the browser using TensorFlow.js. It provides a library of ready-to-use models for computer vision, audio classification, and natural language processing tasks. The suite includes specialized tools for transforming Python-based Keras models into JSON formats compatible with web environments. It enables the deployment of these models by fetching architectures and weight shards via HTTP for client-side execution. The project covers a broad range of capabilities

jsQR is a JavaScript library for locating and decoding QR codes directly within a browser or Node.js environment. It functions as a client-side scanner and decoder that extracts text and data from images using pixel arrays, removing the need for server-side processing. The library provides tools for QR code localization to identify the exact coordinates of corners and alignment patterns within a larger image. It uses image pattern recognition to isolate the QR code and extract its encoded content. Its internal operations cover binarization-based thresholding, geometric pattern localization,

Background Removal JS is a client-side neural matting library that runs a lightweight AI model directly in the browser to compute alpha mattes, removing image backgrounds without uploading any data to a server. It functions as a browser-based image background removal SDK and client-side image matting library, keeping all processing on the user's device to eliminate server costs and preserve privacy. The project provides a cross-platform creative editing engine that produces identical image and video output across web, mobile, desktop, and server environments from a single codebase. It offers

exif-js is a JavaScript library for extracting image metadata directly in the browser. It functions as a binary image buffer parser that reads raw image bytes to retrieve technical details without requiring files to be uploaded to a server. The library parses data following the EXIF and IPTC standards to extract camera settings, timestamps, GPS coordinates, captions, keywords, and copyright information. It utilizes typed arrays and offset-based traversal to navigate image structures and map numeric identifiers to human-readable tags.

RecordRTC is a JavaScript library for capturing audio, video, and screen streams within the browser. It functions as a media recording tool that enables the capture of media streams and their export as playable files. The library provides specialized utilities for recording desktop windows or entire screens and capturing 2D or 3D animations from canvas elements. It handles these tasks by interfacing with browser media stream objects to save activity as video files.

omni-tools is a browser-based utility suite that provides client-side tools for manipulating PDFs, media files, and data formats. It functions as a collection of web-based processors and calculation engines that execute directly within the browser without requiring server-side processing. The suite includes a client-side PDF editor for merging, splitting, and reorganizing document structures, and a web-based media processor for resizing, trimming, and converting image and video files. It also features a data format converter that transforms structured information between JSON, CSV, and XML fo

This repository is for the legacy Mixed Reality Toolkit (MRTK) v2. For the latest version of the MRTK please visit https://github.com/MixedRealityToolkit/MixedRealityToolkit-Unity

This project is a jQuery-based file upload widget that provides a browser interface for selecting and transferring files to a server. It functions as a chunked file uploader and a resumable upload client, enabling the transfer of large files by splitting them into smaller segments or continuing interrupted uploads from the last successfully sent byte. The tool includes a client-side image processor for validating, resizing, and generating previews of images before transmission. It also serves as a cross-domain upload tool, using various transport mechanisms to send files to different domains

ffmpeg.wasm is a browser-based multimedia processing engine that brings the capabilities of the FFmpeg library directly to the client environment. By utilizing WebAssembly, it enables audio and video transcoding, format conversion, and stream recording to occur entirely within the browser without requiring server-side infrastructure. The library distinguishes itself by executing resource-intensive media tasks in background threads, ensuring that the main user interface remains responsive during complex operations. It manages data through an isolated, in-memory virtual file system, allowing fo

Squoosh is a browser-based image optimizer that compresses and converts image files directly within the local environment. By performing all operations on the user device, it eliminates the need for server-side processing, ensuring that sensitive data remains private and reducing network latency. The tool utilizes a collection of high-performance image codecs compiled via WebAssembly to provide professional-grade file optimization and format conversion. To maintain interface responsiveness during resource-intensive tasks, the application offloads image manipulation to background threads and u

SponsorBlock is a community-driven platform designed to automate the skipping of specific video segments, such as sponsorships and intros, during playback. It functions as a crowdsourced database where users submit and categorize video timestamps, which are then retrieved by client-side extensions to intercept and skip unwanted content. The system distinguishes itself through a reputation-weighted consensus model, where community voting and administrative moderation ensure the accuracy of submitted data. To protect user privacy, the platform utilizes hash-based retrieval, allowing clients to

Dropzone is a JavaScript file upload library that provides a browser-based interface for capturing dropped files and transmitting them to a server. It functions as a multipart upload client, splitting large files into smaller chunks to increase transmission reliability and bypass server size limits. The library includes a client-side image processor capable of resizing images and correcting photo orientation in the browser before they are sent to a server. It generates image thumbnails and visual previews immediately after selection, and can render previews for files already hosted on a serve

Supersplat is a web-based suite for editing, optimizing, and viewing 3D Gaussian Splats. It functions as a visual editor for cleaning scenes and a converter for transforming splat data between different file formats to ensure compatibility across tools. The project provides specialized capabilities for 3D scene curation, allowing the configuration of cameras, animations, and annotations. It includes a data optimizer that compresses large spatial datasets into streamed formats to improve loading speeds on web pages. The system further supports the publishing of processed assets to a public ga