Algorithms behind a body measurement app built on smartphones from supplied color-coded garments for online apparel purchasing platforms
Online clothing retailers experience high return rates due to fit and size issues. Most consumers don't know their sizes when shopping online and have significant difficulty attempting to take their own measurements. The proposed body measurement system measures body dimensions from two-dimensional images captured by a user. Measurements of the individual are taken from captured images while wearing one or more coded dimensioning garments designed by the team, with markings at specific locations that can be aligned with characteristic body features. Artificial neural network models are used to track these markings and extract key body dimensions. Exported body data can be plugged into applications such as size selection, clothing customization, fitness monitoring, and training/coaching. Such a solution can be adopted by smart device users easily and has the potential to reduce the return rate for online apparel purchasing platforms and e-retailers.