Canvas Idea

Developed and tested a proof-of-concept for an interactive AI-powered shop window, confirming the technical hypothesis for a unique customer experience.

Canvas Challenge

The core problem for any retail spot is the 'Curse of the Anonymous Shopper.' A shop window is a silent tool that can't establish a personal connection. The challenge was to turn a passive passerby into an engaged guest before they even walked in the door, creating a 'wow effect' and an instant emotional connection.

Stroke Solution

The key decision—the 'one precise stroke'—was the development of a 'Living Dialogue with a Shop Window.' This system combined two technologies: 1) AI Recognition: An integrated camera with a neural network instantly and discreetly identified a person's gender and approximate age. 2) Personalized AI Voice: The system then synthesized a unique voice greeting relevant to the recognized demographic.

Value Metrics

• Technical Hypothesis Confirmed: The prototype proved that the combination of AI recognition and speech synthesis could work in a real-world retail environment to create a unique user experience.
• Invaluable Insights Gained: The local experiment allowed us to gather data on how real people react to this type of interaction, identifying potential issues and growth points for future projects.
• Demonstrated Innovative Potential: The project became a powerful internal case study, showcasing our team's ability to rapidly build and test bold, unconventional solutions at the intersection of technologies.

Tech Stack

• Computer Vision: Python with OpenCV for video stream processing
• AI/ML: PyTorch with a pre-trained model (e.g., MobileNet-SSD) for real-time age and gender detection
• Speech Synthesis: Google Cloud Text-to-Speech API for high-quality voice generation
• Hardware: Single-board computer (e.g., Raspberry Pi 4/5) with a USB camera and speakers
• Orchestration: A single Python script to manage the camera input, AI inference, and API calls

“The question was simple: 'Can we make a shop window talk to people?' We took a camera, a recognition neural network, and a speech synthesizer. For one week, an ordinary window in an ordinary bakery became a small portal to the future. It was pure R&D, an experiment for its own sake. It's projects like these, unafraid to be just a bold idea, that push technology forward.”— The Development Team