Zalando has teamed up with Google on a project allowing users to create their own fashion designs using machine learning, which debuts at fashion festival Bread & Butter in Berlin today
The German etailer Zalando has worked with Google Zoo and digital studio Stinkdigital on a design engine that incorporates the style preferences of more than 600 fashion influencers. The influencers answered a questionnaire that linked their fashion preferences with music, colours, patterns and emotions, which was then combined with data insights from Google and Zalando.
Visitors to the new consumer-facing Bread & Butter, which runs 2-4 September, are able to test the technology, which has been dubbed “Project Muze”, today. They will be asked to answer three questions on the website and the design engine will translate the answers into a virtual design tailored to their responses and preferences. Three looks have been made into physical designs via Zalando’s brand development subsidiary ZLabels to showcase the technology.
“For Bread & Butter, we wanted to experiment with something totally new involving fashion and technology,” said Achim Rietze, strategy lead at Google Zoo for EMEA. “Using Google’s open-source platforms TensorFlow and Cloud Vision API, we explored the creative use of machine learning. This is a trial but we think it is definitely something that will develop further in future.”
Julia Kümmel, global digital communications lead at Zalando, said: “Consumers want to get more involved in what they see and buy than ever before and that will only accelerate with the rise of social media. It’s all about creating individual and emotional experiences.”
Reiner Kraft, Zalando’s vice president of engineering for search and personalisation, said artificial intelligence (AI) is a big focus of research for the firm, alongside virtual reality technologies.
“Shopping is a very personal process and in physical stores customers can utilise the shop assistant’s knowledge to find out what they need and want. We want to explore how we can combine the sought-after qualities of the online shopping experience with these personal touches. We can use AI to revolutionise this process so that years from now, you’re buying fashion in a completely different way.”
Kraft also believes AI could be used in another way – with your smartphone you could take a picture of someone walking down the street, and an app could deduce what items they’re wearing and where to order them online.
“It could even match these items to your own style profile: whether or not they’re your taste, if each item is suited to you, whether you should consider different styles,” he said. “It could ultimately give you taste suggestions.”
Roland Vollgraf, head research scientist at Zalando, said: “Away from the customer-facing side, AI could also be used as part of intelligent systems in operational processes, such as warehouse logistics. For example, steering the load in the different operations is a rather complex dynamical task but AI systems could be used in the future for really optimal steering efficiency.”