Vision-first Computer Use Agent engine.
Anthropic provider for weaveIntel — chat, tool use, vision, PDF, extended thinking, citations, batches, computer use
PoseTracker is a real-time pose estimation SDK and API for fitness and rehab apps. It runs MoveNet on-device on React Native, iOS, Android and the web. Keypoints are free. An optional API key adds rep counting, joint angles and a form score.
React Native MediaPipe Pose Detection with GPU acceleration, jump detection, and high-performance analysis
An easy-to-use drop-in library that wraps around TensorFlow.js BodyPix. Using it is as simple as passing a video reference to the `JerseyDetector` component.
This is simple face detection using face-api.js and tensorflow.js
Run modern deep learning models in the browser.
BBoxJS is a light-weight Javascript solution for marking bounding boxes on an image
Run AI models anywhere.
Blazing-fast, type-safe and lightweight PP-DocLayoutV2 library for Web, Node.js/Deno/Bun. Powered by ONNX Runtime and PaddlePaddle, it delivers state-of-the-art document layout detection with unparalleled speed and accuracy. Perfect for document analysis,
DSH 桌面自动化插件:无障碍树优先的观察-动作回路,30 个标准电脑控制工具(元素/坐标双目标、auto/a11y/event 策略),写后作废/禁重放/持久急停等安全语义,对话卡片+悬浮面板+环境自检设置页,分块识图,JSONL 审计与 pid 白名单。默认关闭。
Computer-vision and OCR-based element assertions for Playwright — local, fast, no GPT Vision required
Provider-independent Node.js classification, spatial localization, and caller-owned mask-based object extraction
A robust and optimized JavaScript library for integrating Google's Teachable Machine models, supporting various image sources and providing efficient classification capabilities.
PoseTracker is a real-time pose estimation SDK and API for fitness and rehab apps. It runs MoveNet on-device on React Native, iOS, Android and the web. Keypoints are free. An optional API key adds rep counting, joint angles and a form score.
Refactored PaddleOCR JavaScript wrapper - strict TS, model sharing, OCR/Table/Formula/Barcode/Layout for browser and Node.js
Production-ready React Native library for intelligent image analysis using Google ML Kit (on-device)
Lightweight, type-safe document image correction for Node.js, Bun, and browsers. Provides orientation classification, text unwarping (UVDoc), and text line orientation correction.
OpenCV WASM Javascript package for video tracking and optical flow including contrib. Slim and more powerful than Official JS.
Intuned SDK computer-use agent: drive a desktop through xdotool with a vision-capable model
Overlay the Cordova webview onto a live video stream
AI-powered image captioning for Vue.js applications using Hugging Face BLIP models
Asynchronous OpenCV 3.x nodejs bindings with JavaScript and TypeScript API.
Microsoft Cognitive Services
Use Roboflow Train computer vision models for inference on images in node.js
Smart screen lock that automatically secures your computer when you walk away.
A Vue.js canvas component designed for machine learning annotation tasks and AI tool development
CROAK - Computer Recognition Orchestration Agent Kit. Agentic framework for object detection model development.
Native OpenCV-based template matching module for Node.js, replacing opencv4nodejs-m1.
Official Node.js / TypeScript client for the SightRadar face recognition API
Open Source Computer Vision Library compiled into a npm package
A simple but comprehensive NodeJS library for Microsoft Computer Vision API
Face detection and recognition library using MediaPipe with helpful React hooks and components
Cloudmersive API Client - Validation, Data and Document Conversion, Computer Vision, Natural Language Processing, and many more APIs.
Computer Vision SDK for Vue
MediaPipe frame processor for VisionCamera in React Native
AI-powered 360° product view generator from single images using OpenAI DALL-E and Stability AI
Models, workflows and runtimes for in-browser perception (scene keyframes, segmentation), for TraffickCam web projects.
Asynchronous OpenCV 3.x nodejs bindings with JavaScript and TypeScript API.
On-device computer vision & VLM multimodal inference framework utilizing device resources for Android Termux & ARM64
Persistent memory, visual session tracking, and RL trajectory capture for AR/XR — the shared brain for every headset
Run modern deep learning models in Node.js.
Asynchronous OpenCV 4.x nodejs bindings with JavaScript and TypeScript API.
GPU-accelerated Computer Vision for JavaScript
Computer Vision Library
Intelligent video analysis powered by OpenAI Vision API. Extract frames, analyze content, search videos semantically, and estimate token costs with FFmpeg and GPT-4/GPT-5 models.