Rekindle

Inspiration

We recognized how challenging it can be to find alternatives for our favorite clothing items, especially when they're discontinued. Our goal was to reignite the excitement of rediscovering your personal style by leveraging the power of AI.

What it does

Rekindle lets you upload a photo of a clothing item you like and uses AI-powered visual search to match the item with similar alternatives. You can customize your preferences and instantly get personalized recommendations.

How we built it

We combined modern AI techniques with a robust database of products. Using visual search algorithms and style-matching techniques, our backend integrates tools like CLIP from Hugging Face and Flask for seamless user experience.

Challenges we ran into

• Ensuring accurate visual similarity with varied image quality

• Leveraging advanced web scraping techniques to systematically extract, process, and curate our data

• Integrating diverse data sources while maintaining a smooth user experience

Accomplishments that we're proud of

• Successfully matching discontinued items with high-quality alternatives

• Creating an intuitive, user-friendly interface that works seamlessly across devices

• Building a scalable system that leverages cutting-edge AI technology

What we learned

We developed a strong understanding of the details of visual similarity search and learned the importance of personalization in e-commerce. We also learned a lot about the challenges of team collaboration, as we frequently dealt with a variety of merge conflicts.

What's next for Rekindle

We plan to expand our product catalog, enhance our recommendation algorithms, and add features like community-driven reviews and style boards, making it even easier to rediscover and reinvent your wardrobe.

Built With

• css
• flask
• git
• html
• javascript
• mongodb
• namecheap
• python
• react
• render
• typescript