MACRO-VISION

Inspiration

The abundance of food we were enjoying during the hackathon sparked a curiosity about the nutritional value of what we were consuming. This inspired us to create a solution that helps others understand their food intake more clearly.

What It Does

Macro-Vision is a web application hosted on a Raspberry Pi that allows users to upload images of their meals. The application compresses and processes these images to provide a detailed nutritional analysis of the detected food items.

How We Built It

We integrated a food detection API to identify various foods from user-submitted images and retrieve nutritional data. A Raspberry Pi 4 serves as the backbone, hosting the website and storing the images for processing. By using Python and various libraries, we streamlined the interaction between image processing, data requests, and the presentation of nutritional facts.

Challenges We Ran Into

Our biggest challenge was the realization that one day was not enough to implement everything we envisioned. Initially, we aimed to create an ASL (American Sign Language) Translator and successfully trained a model for ASL alphabet recognition. However, integrating natural language processing for full sentence translation proved to be beyond our experience within the hackathon's timeframe.

Accomplishments That We're Proud Of

1. Successfully trained our first machine learning models.
2. Implemented and integrated APIs efficiently.
3. Most importantly, we made lifelong memories and friends through this hackathon experience.

What We Learned

We learned the importance of having a solid mission plan for hackathons, which includes conducting risk analysis and having an early pivot strategy. This experience has highlighted the value of preparation and adaptability in project development.

What's Next for Macro-Vision: Nutrition Analysis

We view Macro-Vision as a project in its infancy, one that we plan to continue nurturing. We aim to expand its capabilities, refine the nutritional analysis, and potentially incorporate new features that could make it even more user-friendly and informative.

Built With

• https://vision.foodvisor.io/platform/quickstart/
• python