EcoSnap

EcoSnap is a game that uses AI to turn your walks into a way to help the environment. You can scan trash, find better ways to reuse things, earn points, and help clean up your neighborhood.

Comment

EcoSnap

Inspiration

Waste is everywhere in cities, but most people pass by it every day without feeling like they can do anything about it.

During our morning runs and walks, we noticed that millions of phones see trash before city officials do, but that information is never really recorded.

EcoSnap started with a simple idea: *What if everyday movement could help clean up cities? *

We wanted to take frustration and turn it into action, and use AI as a helpful tool for the community, not just as a cool demo on a screen.

What it does

EcoSnap is an AI-powered cleanup game.

  • It uses the phone camera to find litter while you're walking or running
  • AI sorts out what kind of waste it is (like plastic, food scraps, metal, etc.)
  • It gives safe and useful ideas for reusing or re-purposing the waste
  • It gives you points and streaks to keep you motivated
  • It updates a real-time score for how clean your neighborhood is

In short: *move your body, scan waste, help your city. *

How we built it

We used a thin app, thick intelligence approach to build EcoSnap.

Core components:

  • Flutter for mobile and web (so we can use one codebase)
  • Gemini 3 multimodal AI for understanding and analyzing images
  • Cloud backend for classifying waste, keeping scores, and showing maps
  • Gamification engine for points, cleaning zones, and leaderboards

AI flow:

Camera Image → Gemini 3 → Waste Classification → Action Suggestion → Points + Map Update

We treat neighborhood cleanliness as a dynamic score:

$$ ZoneScore = \sum(DetectedWaste) - \sum(VerifiedCleanup) $$

This makes the impact clear and measurable.

Challenges we ran into

  • Real-world images are hard to work with (like bad lighting or only parts of an object)
  • Preventing fake scans without making the app annoying to use
  • Balancing battery life with scan accuracy
  • Making safe, practical, and local AI suggestions
  • Turning environmental action into a fun game, not something boring

Each challenge pushed us to think smarter, not just harder.

Accomplishments we're proud of

  • Built a real-world AI app, not just a test in a lab
  • Connected fitness, sustainability, and AI in one app
  • Designed a system where every scan helps the community
  • Created a concept that can grow from individuals to entire cities
  • Made the experience simple, fast, and easy to use

What we learned

  • AI is most useful when it helps people take action
  • Gamification can change behavior without forcing it
  • A clean user experience is better than complicated features
  • Environmental problems are also data problems
  • Simplicity works best in real-world situations

We learned that impact matters more than being new or fancy.

What's next for EcoSnap

*ð Predictive mapping of waste hotspots *ðï¸ City dashboards for planning cleanups *ð School and community clean-up challenges *ð Expansion to more cities and regions *ð±Â Scanning waste offline with AI on the phone

Our long-term vision is to make EcoSnap part of a city’s digital nervous system.

*If a city can see its waste in real time through its citizens, does cleanliness stop being a responsibility—and become a shared game? *

Share this project:

Updates