Skip to content

README.md

Latest commit

 

History

History
99 lines (72 loc) · 3.62 KB

README.md

File metadata and controls

99 lines (72 loc) · 3.62 KB

AI Photo Manager

Your memories, organized.

Stop digging through endless folders and cryptic filenames. AI Photo Manager transforms your local collection of photos and videos into a smart, beautiful, and searchable library—running entirely on your own computer. No cloud uploads, no privacy concerns, just your memories.

It organizes your life into three simple views:

👥 People

Forgot where you saved those photos of your best friend? The built-in AI automatically detects faces and groups them together.

  • Identify unknown faces with a click.
  • Merge duplicates easily.
  • Search your entire library just by clicking a person's face.

🌍 Location

Relive your journeys on an interactive world map.

  • See the World: Your photos are automatically grouped into dynamic clusters, showing you exactly where you've been.
  • Travel Timeline: Jump from Paris to Tokyo instantly by browsing your travel history chronologically directly on the map.

📸 Albums

Browse your folders as beautiful, high-performance visual grids.

  • Seamless Playback: Videos play instantly right alongside your photos.
  • Smart Metadata: Your edits and AI data are stored locally in the folder, so your library is always portable and safe.

🛠️ Prerequisites

  • Python 3.8+
  • Visual C++ Redistributable (Required for OpenCV/InsightFace on Windows)

📦 Installation

  1. Clone or Download this repository.

  2. Install Python Dependencies: Create a requirements.txt file (or just run the command below) with the following packages:

    flask
opencv-python
insightface
onnxruntime
pillow
scikit-learn
numpy

    Run the installation:

    pip install -r requirements.txt

    Note: If you don't have a dedicated GPU, onnxruntime (CPU) is sufficient. For faster processing with NVIDIA cards, install onnxruntime-gpu.

  3. Setup Credentials: The system will automatically generate a settings.json file on the first run, but you can create it manually to set your password immediately:

    Create a file named settings.json in the root directory:

    {
   "users": {
      "admin": "your_password_here"
   }
}

🖥️ Usage

  1. Start the Server: Run the application by pointing it to your main photo directory.

    python app.py --root-photo-dir "C:/Path/To/My/Photos"

  2. Access the Web Interface: Open your browser and navigate to: http://localhost:5000

  3. Login: Use the username admin and the password you set in settings.json (default is pass).

  4. Scan Library: Click the "🔄 Refresh" button in the sidebar to start scanning photos, detecting faces, and extracting GPS data.

📂 Data Structure

This application does not use a central database for photo metadata. Instead, it uses Sidecar Files:

  • profile.json: Created inside each photo subdirectory. Contains face encodings, bounding boxes, and image hashes for that specific folder.
  • people.json: Located in the application root. Maps unique UUIDs to Person Names and Avatars.

🤝 Face Management Workflow

  1. Scan: Wait for the scanner to finish.
  2. Identify: Go to an album or photo. Click on an "Unknown" face (dashed border).
  3. Name: Enter a name to create a new person.
  4. Merge: If the system finds the same person but thinks they are new, click the face, select the existing name from the dropdown, and click "Save" to merge them.

📄 License

MIT License (or whichever license you prefer)