News


An artificial intelligence app that can 'Mirrorsize' apparel

11 November, 2018

Getting an apparel of your choice that fits right is a challenge, especially if it is ordered online.
Studies show that nearly 30 per cent of online users drop out before a purchase, because of size and fit issues. Returns, post an online purchase due to sizing issues, range anywhere between 25 and 40 per cent.
And this is where US-based tech start-up ‘Mirrorsize’ comes in with its artificial intelligence-enabled device agnostic body scanning app.
Launched on a trial basis earlier this month — on both Android and iOS platforms — the app takes 3D body scan to deliver “precise body measurements” on tablet PCs and smartphones.

While an individual user gets the right fit, for an apparel maker or e-tailer it could lead to increased sales, lesser returns and customer loyalty. For custom tailoring outfits, it is a chance to enhance their business opportunities and reach larger audiences.
No wonder then that Arup Chakraborty, Founder and CEO, Mirrorsize, wants to extend his offerings to brands, e-commerce companies and even bespoke players (custom tailoring outfits and start-ups). Currently, individual users and some bespoke players are using the app.
“We are in the process of filing patents. Once through, we will focus on the go to market strategy,” he told BusinessLine.
Mirrorsize has already developed two products: ‘Get Measured’ for custom-tailoring outfits; while ‘Size2Fit’ is targeted towards ready-made apparel makers. More additions in terms of product expansion are being planned.
Getting the idea
Chakraborty admits that he was on the look-out for custom tailoring, but was having fitting issues. Tailors would either ask him to be present in person at their shops or he used to send them his measurements. Size charts across e-tailers would confuse him. Even images used in ‘virtual trial rooms’ were not up to the mark. Thus, began his quest for “a solution”. During this time, he met a couple of professors from IIT-Delhi and, along with them, his company Mirrosize started developing the product.
“Measurements taken will be precise because of the use of AI (artificial intelligence). This apart, we will be a cost-effective solution provider,” Chakraborty said.
Monetisation, licensing
Monetisation plans are being worked at. "We may look at a click-based model for companies and brands. I am not much interested in charging individual users," he added. The company is already in discussions with at least seven to eight brands and custom-tailoring outfits (bespoke tailors) for licensing pacts. This means, an apparel maker will provide Mirrorsize with its brand specifications. A buyer will click on the ‘body measurement’ option on the apparel maker’s website or an online shopping site and get to know what size fits him/her the best.
Mirrorsize is also in talks to raise funds which, according to Chakraborty, should not be more than $5 million, as of now. Funds will be used for setting up a global sales and marketing team and to step up focus on R&D ...

https://www.thehindubusinessline.com/info-tech/an-artificial-intellignce-app-that-mirrorsize/article25469039.ece?utm_campaign=amp_article_share&utm_medium=referral&utm_source=whatsapp.com


Made in India AI creates garment industry sizing solution

11 December, 2018

A survey of 1000 online shoppers by British consumer analysis firm Savvy Marketing in 2016 showed that 63% of buyers returned women’s clothing. It was found that online buying and returns were more in the 18-24 age group. Students in this age group went on to say that they bought a lot of clothes because they did not want to be seen with the same clothes on Facebook everyday. They also admitted ordering different sizes because it was impossible to know which would fit. As a result, they ended up sending back half of what they bought.

Is shape important in fashion wear?
Many of us believe that shape is critical to fashion. But those avid watchers of the history of retail garments would know that the answer was actually a resounding NO. Shape was not important for most mass merchandisers. They who defined consumer street fashion accommodated 3 billion plus women, and an equal number of men in around a dozen sizes. In India, it is less than half a dozen for most brands - XXL, XL, L, M and S. And every one who wanted branded clothing but could not afford customised haute couture, would get themselves lumped in sizes that actually did not fit them.

Worse, each brand had a different measurement for every size. So you could be L for Marks and Spencer and still be XL for Calvin Klein’s. So consumers buying ‘ready mades’ were a confused lot. Especially the millennials who believed that going shopping to a mall was a drag. It was far more satisfying for them to browse and buy where the choice was immensely more -- and you could return at will. All that without having to visit the stores even once. The returns in the fashion segment alone accounts for a quarter of the total returns of online sales. Women’s fashion has the most returns and sixty percent of the returns are due to size. One in four loose fitting dresses like T shirts and loungewear are returned by online shoppers due to size misfit while for body hugging apparel like jeans and tops the returns are as high as 50%.

AI helps develop body shape around 2D images
Garment majors across the world have been grappling with the problem of standardising the neck size and chest size of consumers. Usually, sizes differ every two inches except between size 36 to 40 where the clothing could be with variations for every inch. Now a New Jersey based startup, founded by CA and businessman Arup Chakraborty, along with two professors from IIT Delhi, have created an AI based solution for sizing with accuracy up to half an inch. The company which has set up a research centre at a South Delhi neighbourhood with a dozen engineers is bootstrapping and creating a global solution at a low budget with high end programming skills of Indian techies. The AI solution will help merchandisers and retailers attain garment sizing accuracy below half an inch (around 1 cm) which is unprecedented. The technology is based on the development of a model mesh based on two photos of the individual from an ordinary mobile phone of any consumer. All you have to do is keep the phone straight and upload a front side image and a sideways image onto their app.

It is a challenge to develop a three dimensional shape from a front and a side posture photo of any individual. “How would you know in a simple binary image that these are my hand points not all in the same plane? To find that out we did anthropometric identification.” says Arup Chakraborty. “Human bone structure is pretty predefined and catalogued. So we trained thousands of body sets with bones to get the points accurately. Once we got those defining points, it gave us a perfected linear image. The next problem to solve was defining the circumferential, measure with accuracy since there was no depth”.

Micheal Black is a much awarded global expert on computer vision based in Germany. He uses advanced machine learning and graphics to solve problems in the clothing industry and is called the guru of the mesh. In 2013, he cofounded Body Labs Inc. a startup that developed advanced machine vision and optical flow estimation for 3D human shape and motion analysis. He researched with a team of 50 computer vision experts in New York for two years to conceptualise the mesh out of 2D binary images. The company perfected estimation of body shape to the accuracy of four inches ( 10 cms) as per a paper published by Professor Black. Body Shape was bought by Amazon in 2017 for an undisclosed sum, that was estimated to be between $50 to $70 million as reported by Tech Crunch. His work is no longer available in the internet as he is under a standard non disclosure agreement NDA. Amazon has also reportedly taken a patent on the back end processing of Professor Black’s work.

How binary 2D images acquire the 3D shape
So to discover how artificial intelligence converts two 2D images into a 3 D image with accuracy, this author travelled to IIT Delhi to meet Professor Sudipto Mukherjee, a renowned computer scientist who is also the Dean Faculty. “Arup came to us in August 2017 with an interesting problem to solve. We initially thought it was an ordinary computer science problem where you had an image and you need a measurement. The conventional way is to apply geometry from images. But when we started to work that way, we found it cannot be done due to the limits of the speed of activity and the average phone resource with 1000 x1000 pixel resolution” says Prof. Mukherjee. So the work was split into multiple processes and each process was solved differently.

The team combined the deep learning of artificial intelligence which is very good at finding the gross profile with conventional classical solutions that are great for defining small areas. The entire work was broken down into around 16 to 17 modules. While Professor Mukherjee wrote the algorithms for each module at IIT Delhi, like say ‘for creating an edge’, or for ‘creating a centre line’ or for ‘finding the shortest path between a point and a curve’, the programming and the extensive backend coding was done mostly at the basement office by Prof. Chakraborti and his young band of engineers. Some of the coding work initially was also done at IIT Delhi. It was an amalgamation of each of the processes, some solved by classical methods and some by AI that were cascaded to create the total solution.

As a first step, the binary images taken by the consumer were extracted with great accuracy from the background. But they did not synthesise the measurements directly from the 2D image of the human body. They instead added a model placed on the mesh. Then they used the image of the human body and the image of the model and tried to match the two images. They worked back and forth from the human image to the model image and started stretching and pulling the model to get the right measurement and the final shape in the background. It has taken the startup and the professors around eight months of back breaking work and thousands of hours of coding to improve and optimise the accuracy, speed and efficiency of the programme that brings AI and classical programming together to create a much needed solution for the garment industry.

Will this be one of the first success stories of Made in India in the garment industry commercial applications from the field of Artificial Intelligence? It remains to be seen.

https://www.orfonline.org/expert-speak/made-in-india-ai-creates-garment-industry-sizing-solution-46133/


Mirrorsize uses tech to solve the issue of fit in the fashion industry

Mirrorsize’s technique of extracting the body size of customers accurately has won it deals with several online-first fashion retailers, both in India and in the US
05 March, 2019

If Arup Chakraborty has his way, technology built at the Indian Institute of Technology in Delhi (IIT-D) could soon help global fashion retailers assist consumers find the right fitting trousers or suit through their smartphones.

With retail increasingly going the online way, fit issues are among the highest contributors to returns in the fashion industry, one leading to millions of dollars in lost revenues.

Mirrorsize, a two-year old startup, is finding traction among fashion retailers in India and beyond, having already signed up small online brands such as GetStitched, Ira Soleli and Corpqlo.

Some of our competitors want their customers to buy specialised hardware, some want them to install boxes with an array of sensors inside, but we realised the only way to solve this was to build something that was device agnostic where customers don’t have to spend even a dollar on buying hardware,” Mirrorsize founder and CEO Chakraborty told ET. “Most of our competitors use predictive analytics, we did too until 2017, but then scrapped it because no model can fit 6 billion body types.

The company along with IIT-Delhi professors Subhasis Banerjee and Sudipto Mukherjee extensively trained machine vision and classical algorithms, coupled with some hardcore engineering to measure the body size of a person standing feet away from a smartphone camera.
The mixing and matching of techniques has created a far more accurate, if not cutting edge, way of extracting body sizes. The company has applied for a US patent on the technology.

The problem that Mirrorsize is trying to solve has been one of the most sought after uses of technology in the industry, even leading to multi-million dollar acquisitions and investments by giants in the space, the door for disruption is still wide open. Even Flipkart-owned Myntra had made an acquisition in this space, and after launching the product within its app briefly, shuttered the project owing to inaccurate size readings.

" We’ve used AI where it works very well and have refined it using classical algorithms and that’s what has helped us get so far. This isn’t something we could publish papers on, this is hardcore development work. Arup gave us a target and we just put it together, " said Subhashis Banerjee, Professor of Computer Science and Engineering at IIT-Delhi.

Mirrorsize’s technique of extracting the body size of customers accurately has won it deals with several online-first fashion retailers, both in India and in the US. It’s also gaining traction from large players, with the company being in talks to sign on a large e-commerce major in the country and a few massive offline brands that are looking at its technology as a way to build lean online businesses.

https://tech.economictimes.indiatimes.com/news/technology/mirrorsize-uses-tech-developed-by-iit-d-to-solve-the-issue-of-fit-in-the-fashion-industry/68272693


Budding tech startups that are utilizing the VR and AR space

04 February, 2019

We profile two enterprise tech startups that are using advanced technologies like augmented reality (AR) and virtual reality (VR) to provide solutions in diverse areas

HoloSuit: Learning in virtual worlds
Holosuit is a startup focused on changing the way people learn and interact in the real and virtual world. It has developed a bidirectional, wireless and full-body motion capture suit. The full-immersion sensor-packed jumpsuit allows users to point, touch and feel 3D objects, creating a 4D immersive experience. It contains options for multiple embedded sensors and haptic feedback devices dispersed across both arms, legs, and all 10 fingers.
HoloSuit is a wireless platform and comes with a SDK (software development kit) that provides developers with possibilities of creating real and virtual applications.
It is also compatible with Android, Windows, and iOS, Oculus and similar AR/VR /mixed reality (MR) platforms. “When the user is in the virtual world, they can virtually play and learn piano, golf, fly a plane, learn any skill, be in a virtual chemistry lab or physics lab, learn scuba diving," says Shabir Momin, co-founder, HoloSuit.

Mirrorsize: Fit to size using AR On the face of it, this AR startup isn’t doing anything new. Many companies have worked on AR platforms that measure the human body to help e-commerce brands find the right fit for clothes. However, Arup Chakraborty, founder and CEO, claims Mirrorsize’s technology is more advanced than others. He says his technology can get a person’s measurement with an accuracy of one centimetre. A B2B platform, Mirrorsize uses the camera(s) on smartphones to get these measurements. The company has various solutions, tailored to a merchant’s requirements. GetMeasured is a solution for companies who make clothes according to the customer’s measurements. On the other hand, Size2Fit is a solution for those who sell ready-made garments.
The company also plans to launch a solution called Draping. This will allow clothes to be virtually sewn for a customer, meaning users will actually see themselves wearing the clothes through AR technology. Just over two years in the making, Mirrorsize was formed by Chakraborty, alongside two professors from the Indian Institute of Technology (IIT) Delhi—Subhasis Banerjee and Sudipto Mukherjee.

https://www.livemint.com/technology/tech-news/budding-tech-startups-that-are-utilizing-the-vr-and-ar-space-1549220434769.html


Say bye to ill-fitting clothes with Mirrorsize, get AI to drape the outfit on you before purchase

17 January, 2019

The US-based startup's technology can take a user’s measurements with a smartphone and tablet camera for bespoke outfitters. How often are you satisfied with the fit of the clothes you painstakingly picked online? We have all faced situations where we have had to return the clothes we ordered online because they just don’t fit right, even if they are in our size. Agonised with incorrect fits, 51-year-old Arup Chakraborty started Mirrorsize in December 2018. The startup provides Artificial Intelligence (AI)-enabled software that takes measurements of a human body using cameras in smartphones and tablets.

How it all began
Arup, who has lived in the US for more than two decades, has a weakness for Brooks Brothers shirts. “Once I placed an order for a shirt, but when it was delivered it didn’t fit me.I asked for a replacement and again I met with the same fate. Finally, I decided to drive down to the store in New Jersey to get myself measured by their tailor and then get to know the right size,” says Arup, Founder and CEO of Mirrorsize, who we noticed was wearing a Brooks Brothers shirt while speaking to YourStory at a cafe in Delhi’s Connaught Place. Digging deeper, Arup found that this fitting issue is a multi-billion-dollar problem. “Sometimes, online returns are as high as 55 percent globally, predominantly due to fitting issues,” he adds.

Arup noticed that the existing online shopping system either provides size charts to help consumers find the right sized clothing or some e-tailers use data-driven personalisation platforms that ask users a few questions and run an algorithm that makes size recommendations. This was when he decided to solve this complex problem that all brands face today. Arup comes with a finance background and has worked with multiple product startups in the US as their CFO. He has helped these startups raise funding, launch IPOs, and even get acquired by MNCs.With Mirrorsize as a B2B solution in mind, from mid-2016, Arup started talking to various professors across universities such as Stanford University, IITs, and NITs in India. Finally, Arup roped in two professors, Subhasis Banerjee and Sudipto Mukherjee from IIT Delhi, as technical advisors to spearhead the product development process. This also made Arup come back to India. Mirrorsize is a US-headquartered startup that has set up its research centre and its Indian arm in Delhi.

How does the technology work?

So what does Mirrorsize do?
" We generate a 3D model of the object, in this case a human, on the run-time and deform the 3D model into his or her shape,”
explains Arup, adding that no other company does this till date. Arup claims the technology used by Mirrorsize is even more accurate than that which was developed by Germany-based Michael Black, who is known to be the ‘guru of 3D mesh’. Black developed a similar technology to deform a 3D mesh from two images and claimed accuracy of 8 cm - 10 cm. His company, Bodylabs, was later acquired by Amazon in 2017. “We claim our accuracy to be one cm,” Anup quips.

Mirrorsize has different solutions for different businesses. For instance, its ‘GetMeasured’ software is for bespoke merchants to get precise body measurements of their customers, while ‘Size2Fit’ is for merchants selling readymade apparels, where the product gives precise brand-wise size recommendations to a customer. In the first quarter of 2019, the startup intends to launch ‘Draping’, which will let customers select fabric, style etc, and “we will virtually sew the apparel to drape it on the person real time on his/her mobile to give real-life visualisation,” Anup says.

Arup further explains:
“Think of it this way; when a product is pasted on you, if you move, the product does not move with you, but with draping, it is actually like virtually wearing the apparel,”
he adds. Mirrorsize has filed for patents for the technologies developed by the startup.

Not just fashion
Mirrorsize is essentially a B2B startup providing licence to enterprises to integrate its technology on their mobile and web platform with a click of a button. Arup expects to hit $30 million in global revenue for the FY 2019-20.
Arup is not only eyeing the fashion industry but is also looking at sectors like defence, airline, hospitality, and education, etc, wherever there is a need for uniforms. The company has sealed some deals across these sectors in India, the US, Sweden, the UK, and Italy. In Delhi, Get Stitched, a high-end fashion boutique, will be using the technology.

Competition
There are global companies like Tailorguide, Bodi.me, Styku, and Mtailor that have solutions similar to Mirrorsize. “Many are trying to solve this industry-wide issue of wrong size or bad fitting clothes, but they either charge high or need additional hardware or a specific device,” said Arup.
Google launched Tango, a special smartphone with built-in hardware to get precise body measurements, but users have to spend $1,700 to use it. The product was discontinued by Google this year.
Mirrorsize is device-agnostic and no hardware is needed to use the product. “Any smartphone and tablet having a front camera of at least two megapixels will work,” Arup says. Mirrorsize works for all kinds of clothing, including loose-fitting pieces like salwar-kameez.

India to be R&D hub
The bootstrapped Mirrorsize is looking to raise Series A funding of $5 million by the first half of 2019, and will open a new R&D centre in Bengaluru, and will double the team to 30 by next year.

https://yourstory.com/2019/01/say-bye-ill-fitting-clothes-with-mirrorsize/



Mirrorsize US to launch ‘draping’ solution based on customer’s selection of fabric, style

18 December, 2018

"The Indian arm of NJ-based company, Mirrorsize US Inc, plans to launch a ‘Draping’ product in the next three to four months. “Based on customer’s selection of fabric, style, etc, we will sew the apparel to drape it on the person on his/her mobile to give real life visualisation and enrich customer experience,” says Arup Chakraborty, Founder and CEO, of the company. Also on the cards is style recommendation. This will depend on getting funding from a brand. Style recommendation will be based on customer’s style preferences, “We will use deep learning to recommend styles to a customer by geography,” he explains."

The Indian arm of NJ-based company, Mirrorsize US Inc, plans to launch a ‘Draping’ product in the next three to four months. “Based on customer’s selection of fabric, style, etc, we will sew the apparel to drape it on the person on his/her mobile to give real life visualisation and enrich customer experience,” says Arup Chakraborty, Founder and CEO, of the company. Also on the cards is style recommendation. This will depend on getting funding from a brand. Style recommendation will be based on customer’s style preferences, “We will use deep learning to recommend styles to a customer by geography,” he explains.
The company currently offers MD GetMeasured – for bespoke merchants to get precise body measurement of their customers and MS Size2Fit for merchants selling readymade apparels, “where our product will give precise size recommendation to a customer by brand, by apparel and by apparel category. Shirts are available as regular, slim and casual fit,” adds Chakraborty
Solution to solve sizing issues
Chakraborty points out sometimes online returns are as high as 55 per cent, predominantly due to ill-fit. There is noMirrorsize US to launch draping solution based on customers selection of fabric style 001 technology available to get precise human body measurement which prompted them to introduce a solution that cannot only solve ‘sizing’ and ‘personalisation’ issues but also remove supply chain inefficiencies and help scale enterprises by making business more seamless. “This device requires no additional hardware/investment and works on any smartphone and tablet. Users can get their precise body measurement, real-time in seconds. We create a 3D model on the run-time and deform the 3D model to take the shape of the object,” informs Chakraborty.
Independent of background and light
The solution can work with both cluttered and uncluttered background. It is independent of the side pose and can identify another human. It also works in low light. “We generate 3D model of the object on the run-time and deform the 3D model to the shape of the object,” he adds. Drawing a comparison with Michael Black from MPI Germany, known to be the guru of 3D mesh who developed a similar technology to deform a 3D mesh from two images, Chakraborty notes, “Black formed a company called Bodylabs.com and after R&D for over two years, filled his paper claiming accuracy of 8-10 cm. Bodylabs was acquired by Amazon in 2017. Our accuracy is CM and we’re in the process of filing our paper at IEEE in early 2019,”he explains.
Robust plans ahead
Though users can wear both tight and loose fitting garments, the company prefers them to wear tight fitted cloths for accurate results. “We use Physics, Simulation, FEM, Computer Graphics etc. to drape our user. It’s an enterprise SaaS model, which will allow brands, retailers and institutions to use our product,” Chakraborty avers. The company plans to roll-out GTM strategy in Q1’19 and expects to hit $30 million in global revenue (SaaS) in FY 2019-20 as per US GAAP.

http://www.fashionatingworld.com/new1-2/mirrorsize-us-to-launch-draping-solution-based-on-customer-s-selection-of-fabric-style



Mirrorsize US to launch ‘draping’ solution based on customer’s selection of fabric, style

18 December, 2018

"The Indian arm of NJ-based company, Mirrorsize US Inc, plans to launch a ‘Draping’ product in the next three to four months. “Based on customer’s selection of fabric, style, etc, we will sew the apparel to drape it on the person on his/her mobile to give real life visualisation and enrich customer experience,” says Arup Chakraborty, Founder and CEO, of the company. Also on the cards is style recommendation. This will depend on getting funding from a brand. Style recommendation will be based on customer’s style preferences, “We will use deep learning to recommend styles to a customer by geography,” he explains."

The Indian arm of NJ-based company, Mirrorsize US Inc, plans to launch a ‘Draping’ product in the next three to four months. “Based on customer’s selection of fabric, style, etc, we will sew the apparel to drape it on the person on his/her mobile to give real life visualisation and enrich customer experience,” says Arup Chakraborty, Founder and CEO, of the company. Also on the cards is style recommendation. This will depend on getting funding from a brand. Style recommendation will be based on customer’s style preferences, “We will use deep learning to recommend styles to a customer by geography,” he explains.
The company currently offers MD GetMeasured – for bespoke merchants to get precise body measurement of their customers and MS Size2Fit for merchants selling readymade apparels, “where our product will give precise size recommendation to a customer by brand, by apparel and by apparel category. Shirts are available as regular, slim and casual fit,” adds Chakraborty
Solution to solve sizing issues
Chakraborty points out sometimes online returns are as high as 55 per cent, predominantly due to ill-fit. There is noMirrorsize US to launch draping solution based on customers selection of fabric style 001 technology available to get precise human body measurement which prompted them to introduce a solution that cannot only solve ‘sizing’ and ‘personalisation’ issues but also remove supply chain inefficiencies and help scale enterprises by making business more seamless. “This device requires no additional hardware/investment and works on any smartphone and tablet. Users can get their precise body measurement, real-time in seconds. We create a 3D model on the run-time and deform the 3D model to take the shape of the object,” informs Chakraborty.
Independent of background and light
The solution can work with both cluttered and uncluttered background. It is independent of the side pose and can identify another human. It also works in low light. “We generate 3D model of the object on the run-time and deform the 3D model to the shape of the object,” he adds. Drawing a comparison with Michael Black from MPI Germany, known to be the guru of 3D mesh who developed a similar technology to deform a 3D mesh from two images, Chakraborty notes, “Black formed a company called Bodylabs.com and after R&D for over two years, filled his paper claiming accuracy of 8-10 cm. Bodylabs was acquired by Amazon in 2017. Our accuracy is CM and we’re in the process of filing our paper at IEEE in early 2019,”he explains.
Robust plans ahead
Though users can wear both tight and loose fitting garments, the company prefers them to wear tight fitted cloths for accurate results. “We use Physics, Simulation, FEM, Computer Graphics etc. to drape our user. It’s an enterprise SaaS model, which will allow brands, retailers and institutions to use our product,” Chakraborty avers. The company plans to roll-out GTM strategy in Q1’19 and expects to hit $30 million in global revenue (SaaS) in FY 2019-20 as per US GAAP.

http://www.dfupublications.com/news/mirrorsize-us-to-launch-draping-solution-based-on-customer-s-selection-of-fabric-style



Indian techies tackle sizing issues with AI-based solutions

13 December, 2018

A 2016 survey of 1,000 online shoppers by British consumer analysis firm Savvy Marketing showed around 63 per cent of women buyers returned their clothes. Most of these clothes were bought online and buyers were in the age group 18-24 years. Students in this age group bought a lot of clothes as they did not wish to be seen with the same clothes on Facebook every day. They ordered different sizes as it was impossible to know which would fit. As a result, they ended up sending back half of what they bought.

Women’s fashion sees most returns
Earlier, street fashion accommodated 3 billion plus women, and an equal number of men in around a dozen sizes. In India, it is less than half a dozen for most brands - XXL, XL, L, M and S. Anyone, who wanted branded clothing but could not afford customised haute couture, would get themselves lumped in sizes that actually did not fit them. The returns in the fashion segment alone accounts for a quarter of the total Indian techies tackle sizing issues with AI based solutions 001returns of online sales. Women’s fashion has the most returns and sixty percent of the returns are due to size. One in four loose fitting dresses like T shirts and loungewear are returned by online shoppers due to size misfit while for body hugging apparel like jeans and tops the returns are as high as 50 per cent.
AI based solution for accurate garment sizing
Usually, sizes differ every two inches except between size 36 to 40 where the clothing could be with variations for every inch. A New Jersey based startup, founded by CA and businessman Arup Chakraborty, along with two professors from IIT Delhi, has created an AI based solution for sizing with accuracy up to half an inch. This AI solution will help merchandisers and retailers attain garment sizing accuracy below half an inch (around 1 cm) which is unprecedented. The technology is based on the development of a model mesh based on two photos of the individual from an ordinary mobile phone of any consumer.
Micheal Black, a global expert on computer vision based in Germany, in 2013, cofounded Body Labs Inc. a startup that developed advanced machine uses advanced machine learning and graphics to solve problems in the clothing industry and is called the guru of the vision and optical flow estimation for 3D human shape and motion analysis. The company perfected estimation of body shape to the accuracy of four inches (10 cms) as per a paper published by Professor Black. Body Shape was bought by Amazon in 2017. Amazon has also reportedly taken a patent on the back end processing of Professor Black’s work.
Combining AI with conventional solutions
Black along with Professor Sudipto Mukherjee, a renowned computer scientist, combined the deep learning of artificial intelligence with conventional classical solutions. The entire work was broken down into around 16 to 17 modules. While Professor Mukherjee wrote the algorithms for each module at IIT Delhi, the programming and the extensive backend coding was done by Prof. Chakraborti and his young band of engineers. The coding work was an amalgamation of each of the processes, some solved by classical methods and some by AI that were cascaded to create the total solution.
These instances denote the success of commercial application in the garment industry. How much these solutions will sustain remains to be seen.

http://www.dfupublications.com/news/indian-techies-tackle-sizing-issues-with-ai-based-solutions



Indian entrepreneurs could reduce online garment returns with AI

30 December, 2018

Almost two-thirds of shoppers who bought women’s clothes online in the last six months sent at least one item back, according to a 2016 survey by consumer analysis firm Savvy Marketing.Consumers return clothes for a variety of reasons—second thoughts, incorrect purchases, sizing issues and so forth. And digital shopping is compounding the pain.It’s especially hard to know what fits when shopping online. So consumers often order multiple sizes and designs with the intent to send back those that don’t fit.But returns are aren't cheap. On average, returns cost retailers 10 percent of holiday sales.A New Jersey startup called MirrorSize, along with two professors from the Indian Institute of Technology Delhi, may have an AI sizing solution.They’ve set up a research center in a South Delhi neighborhood with a dozen engineers to help merchandisers and retailers accurately size garments within half an inch.Could this be a global solution for fashion’s sizing problems?

Garment sizing confuses consumers
Consumer ready-made fashion is defined by just a handful of sizes—XXL, XL, L, M, S and XS. And some brands forgo sizes altogether in favor of the “one size fits all” tag.Most brands adhere to less than half a dozen sizes to accommodate more than seven billion women and men.And clothing sizes are essentially meaningless when you consider the differences in size and fit between brands.The ranges below, which represent the median 90 percent of styles out of 10,000 jeans assessed, show an astounding variance in waistband distribution of women’s jeans.Women's size distribution How did this happen? Brands have evolved their sizing to represent their core customers— whether they be senior citizens, teenagers or an age group in between. A “medium” for a fashion brand catering to senior citizens is therefore different than a “medium” for a brand for teenagers.It’s no wonder 41 percent of consumers have no idea what size they are when shopping online and buy multiple sizes with the intention of return.
How AI technology shapes body from 2D images
Arup Chakraborty, startup MirrorSize’s founder, sees artificial intelligence (AI) as the solution to the sizing confusion of garment mass production.The technology he’s developing, alongside his partners and several Indian programmers, is based on the development of a mesh of two photos taken by a customer.The entire process was broken down into about 16 modules. Each module contains algorithms for say, “creating an edge,” or “finding the shortest path between a point and a curve”.Eight months of stretching, coding and rework later, they’ve developed a program that can generate precise measurements from two ordinary 2D images. All consumers have to do is upload a front and a side posture photo of themselves to the startup’s app.The technology doesn’t solve the garment industry’s sizing discrepancies, but it could drastically reduce sizing confusion for consumers.And if the tech scales, it could pave the way for more commercial applications of AI in the garment industry.Follow the link below for more on the Make in India startup’s AI program.
Made in India AI creates garment industry sizing solution – Sandip Sen, Observer Research Foundation

https://www.intouch-quality.com/blog/indian-entrepreneurs-could-reduce-online-garment-returns-with-ai