Shozap

• 

  User can interact with the AI-Powered chatbot to get recommendation for shoes form some special occations

• 

  AI-Powered will recommend shoes for your preferred occasions, and you can also be able to virtually try on the shoe by uploading your pics

• 

  Image before virtual try on

• 

  After virtual try on

• 

  Select a shoe from the recommendations.

• 

  Use the virtual try-on feature to see how the shoe looks in 2D on your feet.

• 

  After virtual try on

Inspiration: The growing demand for interactive online shopping experiences inspired Shoezap. We wanted to blend AI-driven recommendations with virtual try-on technology, allowing users to see how shoes look on them before purchasing. Integrating Web3 technology for cryptocurrency transactions further modernizes the shopping experience.

What it Does Shoezap is an innovative platform that engages users through an AI chatbot for shoe recommendations, matches uploaded images to similar products in the catalog, and allows virtual try-ons using object detection. It also supports decentralized purchasing with Tezos cryptocurrency via Temple Wallet integration.

How We Built It We developed Shoezap using:

Backend: Flask (Python) to handle server-side operations. AI: The Gemini API powers chatbot-based shoe recommendations. Image Matching: OpenCV, NumPy, and PIL for processing user-uploaded shoe images and retrieving similar styles. Virtual Try-On: YOLO for detecting feet, allowing users to overlay shoes on their photos. Blockchain Integration: Web3 integration via Beacon SDK and Temple Wallet to facilitate secure, decentralized Tezos transactions. Challenges We Ran Into AI Integration: Ensuring accurate shoe recommendations via the Gemini API required extensive fine-tuning and testing. Virtual Try-On: Implementing real-time foot detection and shoe overlay using YOLO proved to be computationally challenging. Blockchain: Integrating Web3 functionalities with Tezos required careful coordination with the Beacon SDK and wallet security protocols. Accomplishments That We're Proud Of We successfully developed a seamless AI-powered shoe shopping platform with interactive features like image-based shoe retrieval and virtual try-ons. The integration of blockchain for decentralized transactions is another significant milestone, offering users an innovative way to complete purchases.

What We Learned AI Fine-Tuning: Working with the Gemini API and improving recommendation accuracy taught us valuable lessons in API interaction and machine learning refinement. Blockchain Integration: We learned a great deal about decentralized applications, cryptocurrency wallets, and how to incorporate them into consumer-facing platforms. User Experience: Building an intuitive and engaging user interface is crucial for creating an immersive shopping experience. What's Next for Shoezap Enhanced Virtual Try-On: We plan to improve the realism of the virtual try-on experience using advanced 3D modeling and AR technologies. Mobile App Development: Expanding the platform into mobile apps to reach a broader audience. Expanded Web3 Features: Further integration of NFTs and loyalty rewards on the Tezos blockchain to incentivize engagement.

Built With

• allowing-users-to-overlay-shoes-on-their-photos.-blockchain-integration:-web3-integration-via-beacon-sdk-and-temple-wallet-to-facilitate-secure
• and-pil-for-processing-user-uploaded-shoe-images-and-retrieving-similar-styles.-virtual-try-on:-yolo-for-detecting-feet
• beacon
• flask
• geminiapi
• numpy
• opencv
• pil
• sdk
• temple
• tezos
• wallet
• web3
• yolo