Celebrate Your Worth

An Algorithm to Size You Up

AI-driven measurement apps would take exact measurements to break the cycle of trial-return-overproduction

How many times have you ordered clothes online and then had to return them because the size was not right? Although every online apparel merchant meticulously lists the sizes available for each of their products, getting it right is almost always a trial-and-error affair.

Several measurement standards exist; there are variations in size numbering for American and European markets too. And then there is that grey area of “fit”. Often the outer measurements of two similar products coming from different manufacturers might be identical, yet the fitting would vary – with some areas being tight and others loose – for no explicable reason. “Vanity sizing” is also a bane. It is the practice of labelling clothes a size smaller so that brands mislead customers into thinking they are thinner than they actually are

To add to the conundrum, trial and exchange of clothing also comes with its own risk in these pandemic times.Most responsible sellers – both online and in brick-and-mortar outlets, have temporarily suspended return or exchange of tried-out apparel for fear of contagion.

This is where technology comes to the rescue. MySize, an Israeli company harnessed data gathered from customers to develop a smartphone app that can accurately measure body size, completely without the need for physical intervention. The app stores the dimensions of every person using it are store – thus building up a database of sizes based on real-life measurements. Apparel manufacturers and retailers could access and analyse this database to obtain realistic size parameters. That would help them in making and stocking real-life sizes for real people.

The procedure of measurement is simple too. The users only need to move the phone along their body. The app will then leverage the smartphone’s accelerometer sensors and gyroscopes to calculate the dimensions. MySize used the input data it gathered from customers at different outlets to train an artificial intelligence algorithm to recognize the variety of patterns and features inherent in human body shapes. This input was ploughed back to refine its measurements and achieving far greater accuracy than any other similar app.

Yes, there are other measuring apps too, but unlike the AI-approach taken by MySize,they mostly use the smartphone’s embedded camera for measuring purposes. For example, there’s one from MEEPL, a Swiss company, which uses the phone camera to perform a full scan of the user’s body from all sides and creates a three-dimensional rendering of that person. This 3D body double, created through the user’s measurement profile,is then matched against off-the-rack garment data. The app can then suggest the user the specific sizes to look for at retail outlets – on which it has the data available. Online stores canal so use this app – where the customer’s 3D avatar would be uploaded to a virtual trial room to select items.

Industry experts predict that such apps have the potential to be a game-changer in the retail apparels industry. As such, the industry is under pressure to cut wasteful production habits. A United Nations Economic Commission study reveals, 85% of all textiles produced annually – totalling around 21 billion tons a year –are discarded and end up in landfills.

René Stampfl, vice president for Europe, Middle East and Africa at MEEPL,is confident that technology that can create a unique body profile, can help stop this practice. A media report quotes Stampflas saying: “When the consumer shares their personal body measurements with brands and retailers, clothing customisation is achieved for everyone.”

A similar sentiment is echoed by Billy Pardo, chief production and operations officer at MySize, as he spoke to reporters. He pointedly mentions: “When you know the profile of your customers, you will manufacture less, something that will eventually lead to a more sustainable industry.”

Leave a Reply

Your email address will not be published. Required fields are marked *

*