Maige

A professional desktop image editor built with Tauri 2 and native Rust image processing.

Features

• Non-destructive Editing — Exposure, contrast, highlights, shadows, whites, blacks, temperature, tint, saturation, vibrance
• Real-time Histogram — Live RGB and luminance histogram
• Library Management — Import folders, browse and organize photos
• Album Organization — Create albums to group your photos
• Duplicate Detection — Perceptual hashing (dHash) to find similar images
• People Detection — Face detection via MediaPipe with person grouping
• Semantic Search — Search images by content
• Gallery & Editor Views — Grid gallery with single-image editor and filmstrip
• Native App Menu — Full keyboard shortcuts for all operations
• Lightweight — ~15MB bundle, ~30-50MB memory usage

Technology Stack

Concern	Library
Desktop framework	Tauri 2
UI	React 19 + Tailwind CSS
State management	Zustand 5
Build tool	Vite 7
Database	rusqlite (bundled SQLite)
Image processing	Rust (image crate + rayon)
Face detection	MediaPipe Tasks Vision
Virtualization	react-virtuoso

Getting Started

Prerequisites

• Node.js v18+
• Rust (latest stable)
• Platform-specific Tauri dependencies — see Tauri prerequisites

Installation

git clone https://github.com/dileeppandey/maige.git
cd maige
npm install

Development

npm run tauri:dev

This starts both the Vite dev server and the Tauri backend.

Production Build

npm run tauri:build

Project Structure

maige/
├── src/                        # React frontend
│   ├── components/             # UI components (panels, gallery, viewer, modals)
│   ├── store/                  # Zustand stores (edit, library, UI)
│   ├── processing/             # Client-side image processing (canvas, face detection)
│   ├── hooks/                  # Custom hooks (canvas processor, face detection)
│   ├── bridge.ts               # Tauri API bridge (window.api → invoke())
│   └── utils/                  # Utilities (asset URL conversion)
│
├── src-tauri/                  # Tauri backend (Rust)
│   ├── src/
│   │   ├── main.rs             # App entry, native menu, plugin setup
│   │   ├── commands.rs         # IPC command handlers (39 commands)
│   │   ├── database.rs         # SQLite schema and CRUD operations
│   │   └── image_processor.rs  # Native image processing
│   ├── capabilities/           # Tauri 2 permission grants
│   └── tauri.conf.json         # Tauri configuration
│
├── rust/maige-core/            # Shared Rust image processing library
│   └── src/
│       ├── adjustments.rs      # Light & color adjustments (parallel via rayon)
│       ├── phash.rs            # Perceptual hashing for duplicate detection
│       └── processor.rs        # ImageProcessor: load, process, histogram
│
└── shared/                     # Shared TypeScript types
    └── types.ts

Scripts

Script	Description
npm run tauri:dev	Start Tauri in development mode
npm run tauri:build	Build production app
npm run dev	Vite dev server only (frontend)
npm run build	TypeScript check + Vite build
npm run lint	Run ESLint

Performance

Native Rust image processing provides significant speedups over JavaScript:

Operation	JavaScript	Rust	Speedup
Perceptual Hash	~50ms	~5ms	10x
Image Adjustments	~100ms	~10ms	10x
Histogram	~30ms	~3ms	10x

License

MIT