The Sweden Project

This project is a creative tool that transforms your old Minecraft worlds into a personalized, nostalgic experience. It identifies areas where you've spent the most time, captures images of your creations, and presents them in a beautiful web interface, complete with AI-generated captions and music by C418. It's like a digital scrapbook of your Minecraft adventures.

Why Did I Make This?

Like a lot of kids, I grew up playing Minecraft and its a game that I hold quite dear to my heart. To me, it was more than just a game; it was a world where I made some of my best memories, building, exploring, and laughing with friends. I built this project as a sort of tribute to the game and to help others revisit the adventures they once had in those pixelated landscapes.

This project was built solo during the OSDHack Hackathon where the theme was to build something to reimagine something from your past. I built this solo because lately I've been feeling pretty out of touch with my creative and technical side and It is by all means a labour of love, hope you like it :)

How It Works

The process is broken down into three main steps:

1. Scanning: The scanner.py script analyzes your Minecraft world save and identifies "chunks" where you've spent a significant amount of time. This is based on the InhabitedTime data stored in the NBT data of the world files. It filters out chunks that are too close to each other to avoid redundancy.

2. Rendering: For each identified chunk, the mcrender library is used to generate a high-quality 3D render of the area. This creates a visual snapshot of your creations.

3. Interpretation: The rendered image is then passed to an AI model (interpreter.py) that generates a poetic and emotionally resonant caption for the scene. It also selects a fitting C418 song based on the "vibe" of the image.

4. Web Interface: The final output is a beautiful, interactive web page (index.html) that displays the rendered images, captions, and associated music. You can click on each entry to see a larger view and get lost in the memories.

How to Use

1. Installation:

   • Clone this repository.
   • Install the required Python libraries:

     pip install -r requirements.txt

   • Set up your Gemini API key by creating a .env file with the following content:

     GEMINI_API_KEY=your_api_key
     

2. Execution:

   • Run the main script with the path to your Minecraft world:

     python main.py

   • The script will then process your world and generate the necessary files.

3. Viewing the Results:

   • Once the script has finished, open the index.html file in your web browser to view your personalized Minecraft scrapbook.

Project Structure

• main.py: The main script that orchestrates the entire process.
• scanner.py: Scans the Minecraft world for inhabited chunks.
• interpreter.py: Uses an AI model to generate captions and select music.
• requirements.txt: A list of the Python libraries required for this project.
• index.html: The web interface for displaying the final results.
• demo_map/: A sample Minecraft world for testing purposes.
• renders/: The directory where the rendered images are saved.

Customization

You can customize the behavior of the project by modifying the following:

• INHABITED_THRESHOLD in scanner.py: This value determines the minimum time (in ticks) a player must have spent in a chunk for it to be considered "inhabited."
• MIN_DIST in scanner.py: This value sets the minimum distance (in chunks) between two selected chunks.
• emotion_to_song in interpreter.py: This dictionary maps emotions to C418 songs. You can add or modify these mappings to your liking.
• The prompt in interpreter.py: You can change the prompt to guide the AI in generating different styles of captions.

Future Ideas

• User-Uploaded Worlds: A web interface could be created to allow users to upload their own worlds directly.
• More Music Options: The music selection could be expanded to include other artists or even allow users to choose their own.