This project is a demo developed by Sophie Zhu and Pete Sullivan for FORMATION. It showcases a powerful tool designed to enhance the digital map within the FORMATION app or website by generating comprehensive data for a location or object (called a pin) using computer vision.
- Computer Vision Integration: Utilizes OpenAI's GPT-4 vision model to generate detailed data for any pin from an image uploaded by the user.
- Real-Time Updates: The generated data, including descriptions, visual icons, colors, keywords, and more, is added to FORMATION's database via API calls, updating the pin on the app/website in real time.
- Efficient Processing: Although a local, pretrained computer vision model combined with text-to-vector embedding is functional, it is not used due to the superior speed of the GPT-4 vision model.
- Image Upload: Users upload an image of a location or object (pin) within the FORMATION app or website.
- Data Generation: The image is processed using OpenAI's GPT-4 vision model to generate descriptive data, visual icons, colors, and keywords.
- Database Update: The generated data is sent to FORMATION's database through API calls, updating the pin in real time on the digital map.
This demo is run within FORMATION's servers and is designed to seamlessly integrate with the existing infrastructure of the FORMATION app and website.
Since this demo runs within FORMATION's servers, installation and setup instructions are not provided here. For any development or deployment queries, please contact the project maintainers.
demo.mov
- Sophie Zhu
- Pete Sullivan
This project is licensed under the MIT License. See the LICENSE file for details.
For any questions or support, please contact:
- Sophie Zhu: sophie.zhuyy@gmail.com
- Pete Sullivan: petesullivanis@gmail.com
Thank you for using the FORMATION Pin Data Generator Demo. We hope it enhances your experience with the FORMATION app and website!