Skip to content

A JavaFX desktop application for extracting and managing AI image generation metadata across multiple platforms. Features recursive parsing logic, JSON persistence, and a custom dark theme.

Notifications You must be signed in to change notification settings

erroralex/Metadata-Viewer

Repository files navigation

AI Metadata Viewer & Extractor

Java JavaFX CSS Jackson

A high-performance JavaFX desktop application designed to unify generation metadata across the fragmented AI image generation ecosystem. It provides instant extraction, rapid-fire culling, privacy scrubbing, portable local persistence, and deep-node inspection for professional artists and developers.


📸 Interface

Core Workflow

Extractor Portal Speed Sorter ⚡
Extractor View Speed Sorter
Drag & Drop Extraction & Fullscreen Preview Rapid Organization with Keyboard Shortcuts

Library & Privacy

Favorites Library Metadata Scrubber
Favorites Library Scrubber View
Portable Card-Based Persistence Strip EXIF/PNG chunks for privacy
View Advanced Features
Raw JSON Viewer Save to Favorites
Raw Metadata Save Dialog
Deep Inspection for Complex Graphs Themed Undecorated Dialogs

✨ Key Features

  • Speed Sorting (New): A high-velocity workflow for organizing large datasets.
    • Hotkeys 1-5: Instantly move images to pre-assigned folders.
    • Spacebar Navigation: Rapidly skip through sets.
    • Deep Analysis: One-click fullscreen mode for checking fine details before sorting.
  • Fully Portable: The application now stores all favorites and thumbnails in a local /data directory. Move the folder to a USB drive or another PC, and your library travels with you.
  • Universal Compatibility: Intelligent parsing for ComfyUI (API & Workflow), SwarmUI, A1111, Forge, InvokeAI, NovelAI, and SD-Matrix.
  • Metadata Scrubbing: A dedicated view to strip all hidden metadata (Prompts, Workflow, EXIF) and export clean images for safe sharing.
  • Smart Parsing Engine:
    • Content-Aware Detection: Distinguishes between API execution blocks and visual workflow graphs to prevent "N/A" errors.
    • Deep Recursion: Identifies custom nodes (e.g., Power LoRA Loader, Qwen), resolution inputs, and nested JSON structures.
    • Physical Fallback: Reads physical file headers to guarantee valid image dimensions even when metadata is missing or malformed.
  • Interactive UI:
    • Fullscreen Preview: Click any thumbnail (Extractor, Sorter, or Scrubber) for a modal, high-res inspection view.
    • Raw Inspector: Debug non-standard outputs with a syntax-highlighted JSON viewer.
  • Lightweight Performance: Programmatic JavaFX (No FXML) ensures near-instant launch times and zero-lag image processing.

🛠️ Technical Architecture

The application implements a Model-View-Service (MVS) architecture to decouple business logic from the interface.

  • Singleton Pattern: Thread-safe global access to image registries and persistent views.
  • Portable Persistence: Custom JSON serialization logic that maintains relative paths for a self-contained environment.
  • Heuristic Strategy Pattern: Adaptive parsing strategies that score metadata chunks to select the most relevant generation data.
  • Reactive Binding: JavaFX properties ensure real-time UI updates and responsive text wrapping.
  • Technology Stack: Java 8 (Liberica JDK Full), Jackson (JSON Serialization), Metadata Extractor (Drew Noakes), Ikonli (FontAwesome).

🚀 Getting Started

Download Portable Zip


📜 License

Distributed under the MIT License. Free for personal and commercial use.


💖 Support the Project

If the AI Metadata Viewer has streamlined your workflow, consider supporting its ongoing development. Your contributions help maintain compatibility with new AI platforms and node structures.

GitHub Sponsors Ko-fi


Developed by
Alexander Nilsson Logo
Copyright (c) 2025 Alexander Nilsson

About

A JavaFX desktop application for extracting and managing AI image generation metadata across multiple platforms. Features recursive parsing logic, JSON persistence, and a custom dark theme.

Topics

Resources

Stars

Watchers

Forks

Sponsor this project

Packages

No packages published