This project addresses this challenge head-on by introducing an AI-powered digital fitting room. By allowing customers to virtually try on garments and see exactly how they’ll look before making a purchase, we aim to reduce return rates, boost customer confidence, and create a more sustainable, efficient model for online fashion retail.”
Fedora Cabal Wallis.
E-commerce in Mexico has experienced considerable growth, generating a value of $658.3 billion Mexican pesos in 2023, according to the Mexican Association of Online Sales (AMVO). However, this rapid adoption has been accompanied by increased consumer demands, particularly regarding product returns.
According to Tranciti, 40% of consumers return items they purchase online, a phenomenon that negatively impacts the financial health of businesses. Returns in e-commerce can account for a cost of up to 15% of the total sales value. This high cost is due to the operational expenses involved in processing, transportation, and necessary inventory adjustments.
—
Prioritizing efficiency in time and cost, we will implement an integrated tool within the e-commerce platform. This will guide the user to input their body measurements manually. This data will be validated by our AI engine, which will cross-reference it with the garment measurement chart to provide a high-precision size recommendation, significantly reducing purchase uncertainty.
Once the sizing recommendation is established, the process will evolve to include the capture of a user photograph. This image will be processed by AI to generate a visual simulation (virtual try-on) that shows the customer how the garment will look before they purchase it. This final step will ensure satisfaction and maximize confidence.
The second integration aims to maximize user speed. We will use advanced AI (supported by technologies like MediaPipe or DensePose) to create a virtual measurement engine.
Via a simple modal, the customer will only have to input their height. This single piece of data will serve as the base for the AI to accurately extrapolate their full body proportions (waist, chest, etc.).
The system will automatically cross-reference these precise measurements with the clothing inventory, offering the correct size instantly. This eliminates the confusion of size guides and the tedium of manual measurement, accelerating conversion and guaranteeing accuracy.