Fashion is not something that comes naturally to a lot of us. This is why we have the fashion industry guiding us about the latest trends, what not to wear, what definitely not to wear and what not to wear with what. In order to bring a certain calm to the chaos of keeping up with these ever-changing fashion trends, global ecommerce superpower Amazon is currently working on replacing some burden of stylists and designers with AI algorithms. Yes, AI for fashion. Sounds fashionably interesting, right?
Reportedly, Amazon is in the midst of meticulously working on several machine-learning systems that if end up working out perfectly can help not only in spotting and reacting to latest fashion trends, but also shape one's own. The AI systems can help Amazon and several other companies in improving their trend tracking activities in other areas of retail, for example, churning out recommendations based on product spotting on social-media postings. The new effort could also help Amazon in expanding its reach in the clothing industry or even dominate it in the near future.
Last week saw Amazon giving us a glimpse of its interest and thought process for the sector by organising workshop on machine learning and fashion. One of the attendees who signed up for the event was, Kavita Bala, a professor at Cornell University. According to her, recent years have seen a whole lot of companies like Amazon investing their time and money in getting a better understanding of how fashion actually develops and grows in the real world. “This is completely changing the industry," she further added in a statement to MIT Technology Review.
The clothing world is currently witnessing a number of retailers making use of popular social networking sites like Facebook, Instagram and Pinterest to keep a track of the quickly changing fashion trends and keep themselves at pace with these changes. For example, there's a Startup called Stitch Fix, which provides personalised recommendations based on a user's social-media activity and preferences.
In the past few years, Amazon has actually worked hard on growing its apparel business. It has invested in developing its own clothing brands, spent money on good-quality photography shots for its products, and debuted its Prime Wardrobe, a novel concept which allows online shoppers to try clothes before actually buying them. It has even developed an app called Echo which gives users feedback on whether a particular outfit works for them or not.
But, one look at all these activities can help one deduce that the company is more incline towards pushing the algorithm agenda into the industry. For instance, recently a group of Israel-based Amazon researchers had successfully invented a machine learning that could deduce whether a particular looks falls in the stylish category by simply analysing a few labels attached to images. The software developed is capable of providing constructive feedback for making some quick adjustments. For the unversed, the machine learning developed is being considered as extremely innovative as usually a computer needs to be fed extensive labelling in order for them to learn from visual information. But, in the social media world, people might not put that many labels.
At Amazon's San Francisco-based research lab, Lab126, a team has been successful in developing an algorithm that is capable of learning about a particular style of fashion simply from images, and then create new items in similar styles right from the beginning. One can say that the algorithm acts as a simple AI fashion designer. They use a cutting-edge tool called a generative adversarial network, or GAN, which were developed by a researcher on the Google Brain team. The GAN consists of two deep neural networks operating in tandem to learn efficiently from raw data. It internalises the properties of a particular style simply by looking at lots of examples, and it can then apply that style to an existing item of clothing. The system might not be as much sophisticated as the fashion industry wants it to be, but it definitely hints towards a revolution in the making.
Both of the aforementioned Amazon efforts were revealed in Amazon's workshop held last week. The event also saw many companies sharing their innovative efforts in the field of fashion and AI. For instance, a team from Indian clothing ecommerce site Myntra demonstrated a program that guesses a person’s correct size for a particular garment from his or her past purchases. Another interesting idea was from a group from the University of Illinois in Urbana-Champaign who showed an algorithm that could help identify fashion-focused social-network accounts.
It seems, AI for fashion is going to become big very soon. Keep watching this space to remain updated about the same.
Reportedly, Amazon is in the midst of meticulously working on several machine-learning systems that if end up working out perfectly can help not only in spotting and reacting to latest fashion trends, but also shape one's own. The AI systems can help Amazon and several other companies in improving their trend tracking activities in other areas of retail, for example, churning out recommendations based on product spotting on social-media postings. The new effort could also help Amazon in expanding its reach in the clothing industry or even dominate it in the near future.
Last week saw Amazon giving us a glimpse of its interest and thought process for the sector by organising workshop on machine learning and fashion. One of the attendees who signed up for the event was, Kavita Bala, a professor at Cornell University. According to her, recent years have seen a whole lot of companies like Amazon investing their time and money in getting a better understanding of how fashion actually develops and grows in the real world. “This is completely changing the industry," she further added in a statement to MIT Technology Review.
The clothing world is currently witnessing a number of retailers making use of popular social networking sites like Facebook, Instagram and Pinterest to keep a track of the quickly changing fashion trends and keep themselves at pace with these changes. For example, there's a Startup called Stitch Fix, which provides personalised recommendations based on a user's social-media activity and preferences.
In the past few years, Amazon has actually worked hard on growing its apparel business. It has invested in developing its own clothing brands, spent money on good-quality photography shots for its products, and debuted its Prime Wardrobe, a novel concept which allows online shoppers to try clothes before actually buying them. It has even developed an app called Echo which gives users feedback on whether a particular outfit works for them or not.
But, one look at all these activities can help one deduce that the company is more incline towards pushing the algorithm agenda into the industry. For instance, recently a group of Israel-based Amazon researchers had successfully invented a machine learning that could deduce whether a particular looks falls in the stylish category by simply analysing a few labels attached to images. The software developed is capable of providing constructive feedback for making some quick adjustments. For the unversed, the machine learning developed is being considered as extremely innovative as usually a computer needs to be fed extensive labelling in order for them to learn from visual information. But, in the social media world, people might not put that many labels.
At Amazon's San Francisco-based research lab, Lab126, a team has been successful in developing an algorithm that is capable of learning about a particular style of fashion simply from images, and then create new items in similar styles right from the beginning. One can say that the algorithm acts as a simple AI fashion designer. They use a cutting-edge tool called a generative adversarial network, or GAN, which were developed by a researcher on the Google Brain team. The GAN consists of two deep neural networks operating in tandem to learn efficiently from raw data. It internalises the properties of a particular style simply by looking at lots of examples, and it can then apply that style to an existing item of clothing. The system might not be as much sophisticated as the fashion industry wants it to be, but it definitely hints towards a revolution in the making.
Both of the aforementioned Amazon efforts were revealed in Amazon's workshop held last week. The event also saw many companies sharing their innovative efforts in the field of fashion and AI. For instance, a team from Indian clothing ecommerce site Myntra demonstrated a program that guesses a person’s correct size for a particular garment from his or her past purchases. Another interesting idea was from a group from the University of Illinois in Urbana-Champaign who showed an algorithm that could help identify fashion-focused social-network accounts.
It seems, AI for fashion is going to become big very soon. Keep watching this space to remain updated about the same.
Advertisements