INDITECH

Inditech is a nice multimodal AI app designed to transform online shopping at ZARA by simply speaking to search for items or use hand gestures to navigate, making shopping interactive, personalized.

Comment

Inspiration

Our goal was to integrate multimodal AI capabilities to enhance the e-commerce experience for ZARA. We recognized the potential of voice and gesture-based navigation to simplify and enrich online shopping, making it more interactive and user-friendly.

What it does

Inditech is a shopping app where you can use your voice to ask for products, once you speak several items that fit your request will appear in the web, and if you like one of them you can check the outfit from more angles by clicking on it. We also implemented a hand gesture detection algorithm so you can interactively move through the app showing thumbs up or thumbs down.

How we built it

The backend of Inditech was developed using Python with a Flask framework to manage server-side requests and data processing. The frontend was constructed using HTML, CSS, and JavaScript to create an engaging and responsive user interface. For the AI components, we utilized OpenCV for real-time gesture recognition, transformers from Hugging Face for processing natural language queries, Fashion-CLIP for fashion-specific image analysis, and OpenAI Whisper for robust voice recognition capabilities.

Challenges we ran into

We encountered several technical challenges, including compatibility issues between different library versions which required extensive troubleshooting. Implementing the YOLO model for gesture recognition also posed significant difficulties, particularly in accurately detecting thumbs-up gestures in diverse lighting conditions. Additionally, maintaining stable connectivity with our web hosting service was challenging.

Accomplishments that we're proud of

The application functions effectively, demonstrating its ability to understand and process user inputs through both voice and gestures accurately. Thishas huge potential to be adopted by companies like Inditex to modernize their online shopping.

What we learned

The project was a significant learning experience in applying AI in a practical, user-focused context. We enhanced our skills in multimodal AI integration, improved our ability to debug complex systems involving various technologies, and learned to better synchronize our team efforts in a dynamic development environment.

What's next for INDITECH

Ultimately, we intend to pilot this system with ZARA to assess its impact on consumer engagement and sales.

Built With

Share this project:

Updates