Visualize the cities you deserve. An AI tool for transforming urban and rural streets into people-first spaces.
- 📒About The Project
- 💚Example Transformations
- 🔧Built With
- 💻Setup
- ✅Tests
- 🗺️Roadmap
- 🤝Contributing
- ✉️Authors
- ⭐Acknowledgments
- 📦Version History
Ginhawa AI helps you visualize the cities and towns you deserve through AI-powered transformation. The name "Ginhawa" is a Filipino word for "relief" or "comfort," reflecting our goal: to help redesign public spaces so they truly serve people.
This tool empowers anyone to reimagine their streets. Upload a photo of any urban or rural space, and watch as AI generates a new vision—one that is greener, more walkable, and designed for human connection.
Our mission is to inspire change by making the potential of our communities tangible and beautiful.
Original street view of FEU Tech, P. Paredes Street, Manila |
💚Transformed vision |
Original street view of Carriedo Street, Quiapo, Manila |
💚Transformed vision |
Original street view of Zapatero Street, Marikina |
💚Transformed vision |
Original street view along Katipunan Avenue, Quezon City |
💚Transformed vision |
- Frontend: React & TypeScript
- AI Models:
- Google Gemini 2.5 Flash for image analysis.
- Google Imagen 3 for image generation.
- Styling: Modern CSS with Custom Properties
To get a local copy up and running, follow these simple steps.
-
Node.js
-
You will need a Google AI API key. You can get one from the Google AI Studio.
This project is set up to run directly in the browser without a complex build process.
-
Clone the repository:
git clone https://github.com/ralphcajipe/ginhawa-ai.git cd ginhawa-ai -
Set up your API Key: Set the
GEMINI_API_KEYin .env.local to your Gemini API key -
Run the app:
npm run dev
Currently, the project does not have an automated testing suite. Manual testing can be performed by:
- Running the application locally.
- Uploading various street-view images (JPEG, PNG).
- Verifying that a transformed image is generated successfully.
- Checking for proper error handling when an image is not uploaded or the API key is invalid.
- Allow users to customize the transformation prompt (e.g., "spanish colonial era style," "add more bike lanes").
- Add a "Share to Social Media" feature.
- Implement user accounts to save a history of transformations.
- Integrate with a mapping service to tag locations.
See the open issues for a full list of proposed features (and known issues).
If you have a suggestion that would make this better, please fork the repo and create a pull request. You can also simply open an issue with the tag "enhancement."
- Ralph Cajipe - Developer - ralphcajipe
- Google AI for the powerful Gemini and Imagen models.
- Inspiration from urbanists and communities advocating for better cities, especially my friend Stephen who encouraged me to build this.
- v1.0.0 (Current)
- Initial release with core functionality: upload an image and generate its potential transformation.