3D&ME | concept design
NTT held a machine learning hackathon that took place at NestGSV in Sillicon Valley, June 2014. I lead design in a team of 5, building a concept connecting three technologies - the Mind Wave, the Cube 3D printer, and machine learning - and placed in the Top 3.
Develop machine learning scenarios, code and data visualization. We have loaded a few public data sets, but you can use your own data sets or any other public data sets. Explain in your concluding presentation the reasons for choosing a specific machine learning tool set and framework for the selected scenario.
Visualization (meaningful representation of the data and results).
Presentation (details of the use case, description of the tool set and the reasons for choosing it and business applicability).
Use of real-time machine learning (for example, Jubatus) earns extra points.
iOS, 3DSys Cube, Jubatas Machine Learning, the Mind Wave Mobile
Proof-of-Concept - Top 3
3D&Me connects the mind to the 3D printer so that you can hold your thoughts in your hand.
The long version:
Take sensory data from the Mobile Mind Wave and use supervised machine learning algorithm to convert to a personalized 3D object for printing.
The Mind Wave Mobile detects certain emotional levels when worn and the output is visualized on the mobile app. Certain combinations are then sent through our machine learning algorithm to the 3D printer. The final product is a tangible memory or thought.
The idea for the app is to introduce and guide the user through the process. The app is design to get the prompting out of the way initially so that it the user can calm the mind without interruption for the Mind Wave to collect sensory data. Then, the data is analyzed by the machine learning algorithm and a set of printable objects are displayed for the user to choose. The chosen model will be sent to the 3D printer through the cloud.
I could not change the colors of the results bars which were brought over from the Mind Wave API so I decided to match the fun colors and use a dark gray to mute out some of the brightness.
As the only designer on the team, I lead the brainstorming and ideation exercises, creating user flows, building product features, connecting business goals, and the designing of the user experience.
We explored different options on concepts to build. We came up with a bunch of ideas centered around smart technology but none took full advantage of the machine learning technology.
So, I had everyone dig into the story we were trying to tell. I asked what are we passionate about? How can we make a difference?
How we came about the concept was completely serendipitous. I had a 3D printer I won from a previous hackathon just sitting around in the office. My team mentor was playing around with the Mind Wave Mobile when the idea struct: why don't we find a way to connect the two using machine learning?
When the idea was proposed, the team immediately woke with excitement. We started studying the Mind Wave Mobile first because we knew the least about it. We found that levels of sensory data, such as aggravation and calmness, could be detected. Eventually, we fell in love with the idea of using machine learning to analyze certain emotional data collected from the Mind Wave Mobile and then matching it with pattern sets we teach it. The pattern sets are associated with a fixed number objects that can then be sent over the cloud to the 3D printer.
We built this project with 2 groups of people in mind. We envisioned our concept to be useful to the visually impaired who would be able to touch and feel their thoughts. We also envisioned our concept to allow seniors to relive their favorite memories.
The morning of the big pitch, we tested the Mind Wave Mobile device on this lovely volunteer - our teammate's little sister. This was the only down time we had to really see the whole process from beginning to end since the engineering was only solidified early that morning. We wanted to make sure that it would go smoothly in the case that we were asked to demo after the presentation.
Each team was given 5 minutes to pitch our concepts and impress the sponsors/judges. One sponsor was so intrigued that he stopped us in the middle of our presentation and asked to demo it on himself right then and there!