Skip to content

WilliamUW/ETHDenver2024

Repository files navigation

🏋️ BeFit

BeFit revolutionizes fitness motivation by blending the spontaneity of BeReal with the accountability and rewards of exercising.

Mission: To make fitness a shared, rewarding journey, leveraging technology to foster a community of support and motivation.

💡 Inspiration

The challenge of maintaining a consistent exercise routine is UNIVERSAL. With the rise of sedentary lifestyles, finding the motivation to exercise has become increasingly difficult for many. BeFit addresses this challenge by:

  • Encouraging spontaneous physical activity through random exercise prompts.
  • Using machine learning to verify exercise completion, ensuring accountability.
  • Rewarding successfully performed exercises with unique NFTs, creating a tangible record of achievements.
  • Integrating financial incentives to further motivate users and strengthen social bonds.

❓ What it does

BeFit is a groundbreaking fitness app that combines the best aspects of social media spontaneity, accountability in fitness, and the gamification of exercise habits. Here's how it works:

  1. Random Exercise Prompts: At random times, BeFit alerts users with a specified exercise to perform.
  2. Machine Learning Verification: Users complete the exercise in front of their camera, and the app uses machine learning to confirm the activity.
  3. NFT Rewards: Upon successful exercise verification, an NFT is minted as a record of achievement.
  4. Social Motivation: Users can view friends' accomplishments, fostering a supportive community.
  5. Financial Incentives: BeFit allows sending USDC tokens to friends as rewards for meeting fitness goals, adding a financial motivation layer.

🚧 How we built it

Technologies Used: Machine Learning for exercise verification, Blockchain for NFT minting, Cryptocurrency (USDC) for financial incentives, Social Media Integration for community building

  1. Machine Learning: Leveraged cutting-edge ML algorithms to accurately verify exercise completion through camera feeds.

  2. Blockchain and NFTs: Utilized blockchain technology to mint unique NFTs for each verified exercise, creating a collectible and motivational system for users.

  3. Cryptocurrency Rewards: Integrated USDC token transactions to enable users to financially reward their friends, encouraging a healthy competition and mutual motivation.

  4. Social Media Integration: Developed features for users to share their achievements and view their friends' progress, promoting a sense of community and support.

Check out our demo and see how BeFit is changing the fitness game!

Human pose estimation for fitness activity detection using PoseNet (implementing MobileNetV1)

posenet_keypoints posenet_keypoints2

https://blog.tensorflow.org/2018/05/real-time-human-pose-estimation-in.html

Pitch

👀 Slide Deck

The Team

team slide page

Instructions

Run in command line at project root:

yarn install (if doesn't work, use npm install)

yarn dev

Deprecated instructions:

copy .env.example as .env