Objective
A recent study highlighted that 74% of in-store shoppers find the checkout process to be the most frustrating aspect of their retail experience. This served as a catalyst for a prominent U.S. retail chain to consider integrating machine learning and frictionless checkout technology in their compact grocery stores.
Approach
To address this challenge, InContext was engaged to provide extensive 3D modeling services for various inventory categories such as shelf-stable, frozen, and select perishable items at one of the retailer's smaller grocery outlets. The project involved acquiring a large volume of product models, capturing essential item attributes, conducting high-quality studio photography, and ultimately delivering over 29,000 3D models. These models were crucial for enhancing object recognition through machine learning, enabling store cameras to accurately identify items in customers' baskets and ensure correct pricing at checkout.
Outcome
InContext successfully executed this project within a tight 9-week timeframe, demonstrating exceptional attention to detail and zero room for error. The delivery of 3D product models was staggered, averaging nearly 3,500 models per week, showcasing an ecosystem capable of delivering both quality and quantity at scale.