Repository Overview
📝 JavaScript • 📄 CC BY 4.0
Repository: thecont1/ngl-storyteller Last Updated: January 1, 2025
ngl v1.0 😜 – Not Gonna Lie!
A creative photo-based storytelling tool that gives you the freedom to reassemble your visual memories into compelling, surrealistic narratives.
⚠️ Fair warning: This app can be quite addictive once you start building your stories! Use responsibly 😊
🍏 What is ngl?
“A fictional world more representative of reality than dull reality itself.”
Think of ngl as a creative partner for visual storytelling. Upload your photos, and the app uses AI to identify the most significant elements within them.
Then, like assembling a collage, you can layer these elements to create new compositions that capture the emotional truth of your experiences—even if they never existed quite this way in reality.
This approach mirrors how human memory actually works: fragmented, layered, and emotionally weighted rather than photographically accurate.
🎨 The Philosophy
This tool emerged from photographer Mahesh Shantaram’s MFA Photography research examining:
- AI and computational photography can reveal deeper truths about our experiences
- ngl builds on the concept of “softimages”—images that exist as software programs, infinitely adaptable and merged with computational processes
- Represents a fundamentally different kind of photograph for our digital age
The project originally explored the gig economy—Mahesh Shantaram worked as a nightly delivery boy on the streets of Bangalore, on an electric bicycle and documented this already-surreal world using night photography and computational imaging.
This evolved into a broader investigation of how AI can help “reassemble a fictional world more representative of reality than dull reality itself.”
🚀 Getting Started
What You’ll Need
- A modern web browser 🌐
- API key for AI services 🔑
- Your favorite photographs 📸
Installation
- Clone or download this repository
- Install dependencies
- Set up your API key
- Run the app
- Open your browser to the URL shown in your terminal (usually http://localhost:3000)
🎯 How to Use
- Set the scene – Select a photograph that lays out the scene where your story will unfold
- Build your story – Upload new images, one by one. The app uses AI to identify the most significant object(s) in them
- Compose your narrative – Take charge of your story by editing the layers, moving around the objects, and even applying readymade styles
- Generate a caption – Add a border, if you wish, for that photo frame look. Then get AI to generate a witty caption for your surreal story
- Save your Story – Multiple export options available
💾 Save Options
- ☘️ Save Raw Data (.ngl): The most efficient way to save your story. Contains all the data needed to pick up from where you left off
- ☘️ Save Project (.png): A static image of your story loaded with all the layers and styles. Drop this into the app to resume
- ☘️ Download Image (.jpg): A plain and static image of your story. Best for sharing on social media
- ☘️ Save as Magritte: AI-powered feature trained on René Magritte’s iconic works
🎨 Save as Magritte
Save as Magritte is an AI-powered feature trained on René Magritte’s iconic works, allowing you to infuse his distinctive surrealism into your own compositions.
The transformation uses 11 carefully selected Magritte paintings as style references, capturing his:
- Oil painting texture 🎨
- Dreamlike compositions 💭
- Philosophical playfulness 🤔
📊 Cost Monitor
To encourage responsible use of AI resources, ngl includes a built-in cost monitor that tracks your API usage in real-time. This transparency helps you stay aware of the computational resources powering your creative experiments.
🎯 Use Cases
- Personal storytelling: Transform your photo memories into surreal narratives 📖
- Artistic exploration: Experiment with AI-assisted visual composition 🎨
- Educational projects: Explore concepts of memory, reality, and digital photography 🎓
- Creative workshops: Use as a tool for photography and digital art classes 👥
⚙️ Technical Details
Built with modern web technologies for creative AI applications.
📄 License
This project is open source and available under the Creative Commons Attribution 4.0 International License (CC BY 4.0).
🙏 Acknowledgments
- The original project idea was developed as part of Mahesh Shantaram’s MFA Photography research at Ulster University in 2021-2023
- Google DeepMind Hackathon - Vibe Code with Gemini 3 Pro in AI Studio
- WikiArt Visual Art Encyclopedia