Unconventional Design Group Quick and Dirty Design – Zeyao Li

 

Last week Gabe, Xiaoyan, Robin and I worked together on improving Nick’s phone problems. Nicolas is one of the research fellows at IMA. From the interview, we observed that Nick’s biggest phone using issue is that he tends to play his phone at night before he goes to bed. He also complains that his phone has a short battery life yet he does not want to carry his extended battery all the time. Besides these, Nick wants to improve his cooking skill and his lifestyle in general. So after finding out the problems, we decided to tackle the main issue, which is nick’s late-night phone use.

When we brainstormed the idea, we talked about maybe we can make an app that stops nick playing his phone at night, however, an app did not solve the battery issue. So we proposed that we wanted to make a combination of hardware and software. At that moment, I suddenly came up with the idea of using an AI robot to solve Nick’s problem. As all the IMA students knew, Nick is well known for his robots, thus I want to have a robot to help him. We named our product “Nick’s box”. It contains a box where Nick can put his phone in to charge and a robot which will remotely control the box. If Nick wants to take the phone out of the box, he needs to get the robot’s permission. The robot will ask Nick why does he want to use his phone. If the reason is appropriate, the box will be unlocked. However, if Nick lies about his use, then the box will be locked for even longer. Robin and I used Lego to build the robot based on Nick’s robot. The design is really personalized for Nick, and we are all happy about it.

Reading Response – Week 2- Zeyao – UnconvetionalDesign

I started my reading from the medium post “Visualizing the essential of design thinking”. The author firstly points out different models and their weakness in showing the essence of design thinking. Then based on the model that exists, he and other people create the more complete and detailed model to explain the design thinking process. It combines the model that he talks about and puts it together. I like the visualization that the author makes however, I do think it is a little bit too complicated to read for the first time. It contains too much information in one graph. What I like about this graph is that it shows the iteration in both micro and macro scopes. It also well separated the actual design process and the thinking part.

Then I jumped to the reading about a map of design practice and research. Because I think I can find some similarities in between maps and visualizations of design. In my option, a map is a way to visualize stuff and to make things clear (like mind map). The map breaks down two totally different mindsets from two groups of people: designers and researchers (which includes sociologists, engineers and more). It also defines the differences between designers and researchers who have an expert mindset or a participatory mindset. The research types map helps me understand our class more clearly. Since it categorizes different types of design research, which some weekly projects will cover. I think a good designer requires the quick switch from different mindsets because the task from each client is different.

“Wicked Problems in design thinking” is a really theoretical article that explains design and design thinking in our lives. The part that I highlight in the article is that it elaborates four areas that design can apply in real life: visual communication; material objects; activities; and design systems (which part of it I learned in last semester). It helps me understand the use of design and the area of design thinking. The article emphasizes that the goal of designer is to tangle specific problems.

Good Design – Zeyao – UnconventionalDesign – Week 2

For the assignment of “Good Design”, I made a book cover for a classic coming of age book “The Perks of Being a Wallflower”. The book was firstly published in 1999. It speaks from a high school freshman’s perspective and tells stories around him. The original book cover has a classic vibe, shows the green background plus a washed-out picture on the right-top corner. The cover has a sans-serif font to show the modern feeling. However, the original book cover was a little bit outdated nowadays. My goal for the redesign is to make it attractive to younger readers. 

The first published version

My graphic design process always starts with creating a mood board to establish my designing style. I wanted the book cover to look more trendy, therefore I decided to go with illustrations and hand-written style. I collected pictures from the Internet and then put it together as a collage. The pictures that I found all conveyed a similar coming-of-age vibe. It showed youth and high school love. 

My mood board for the book cover

For me, good design requires iterations. I don’t think good design only has one shot. It needs to be updated based on people’s feedback. In the beginning, I drew the illustration that I thought could represent the image of the book and provided different layouts for my friends and others. Then they said the flower part was a little bit emptier. Thus, for the final version, I redesigned the illustration as well as added the hand-written title. I chose a sans-serif font for the author to balance out the hand-written and illustration. 

 
The initial version of my book cover design

Final design

iML – Zeyao – Week 3- What Does Machine think of You?

Assignment 1

Develop a simple project with any ml5.js models covered in class

Intro:

I always use a webcam as a mirror or a photo booth. I know I am the one who is showing on the screen. However, does machine think the same thing as I do? In the era of machine learning, the machine can use the webcam to distinguish different items. One of the tools is the image classifier. So I am wondering what will machine think of who I am? Thus, I create a simple sketch to let machine expose my identity.

Process:

I started the assignment by exploring the image classifier first. I was trying to figure out the output it provides, then limiting the range of the result. Then I used the text function to print out the name of the result. However, it turned out not as good as I thought. The text blinked too fast on the screen to read for the user. Then Moon told me to create a class for all the texts and push it all. I limited the texts length default to 20. So when it reaches to 20 texts on the screen, it splices the first index from the array. After the basic function, I started to smooth the data from the result. The first filter is the undefined (which will be shown before the model gets loaded), then I created another filter to filter out the repetitive and consistent name (using a variable to store the previous frame’s name). After that, the data got clearer. I added the graphics layer on top of the canvas so that it won’t be overlapped with the webcam frames.  Once you click the mouse, the canvas gets clean up. I also added the gui to control the texts length and a function to make the webcam stop at the frame that you like.

Implements

Since this is the first assignment, I just explored the basic stuff with one function. For the next step, I want to use Google image api to push the image that relates to the text and lay it next to the webcam frame. Also, the text position is kinda weird right now (not totally random). So I want to change that.

 

github link:  https://github.com/zeyaoli/Interactive-ML/tree/master/Homework-1

Week 2 Case Study Research – BeautyGAN – Zeyao

BeautyGAN is the machine learning model that I mentioned in the first class. It was a perfect example of applying Machine Learning technique to something artistic. Published by 6 researchers, BeautyGAN is an intense-level facial makeup transfer with deep learning generative adversarial network. It transfers make-up style from the makeup dataset to someone who does not wear any makeups. From the paper, it says they “achieve automatic makeup transfer with a dual input/output generative adversarial network” and “achieve instance-level style transfer by successfully applying pixel-level histogram losses on local regions”. It also builds a makeup dataset with 3834 images. 

You can find the paper here

However, what I want to actually talk about is the use of BeautyGAN. I first discovered this machine learning model on Dazed’s Instagram account. Dazed is a British fashion magazine that has a long history. This year, they released their Dazed Beauty Issue 0, the first issue that focuses on the beauty. For the first issue, they got Kylie Jenner, one of the most influential beauty celebrity in the world, to be on the cover. Different from the traditional cover shooting, Dazed Beauty let Beauty AI does Kylie Jenner’s makeup, which brought the BeautyGAN to the public. 

The article explains how would BeautyGAN work in a literal way. “The AI starts out with a data set: 17,000 images pulled from Instagram by the Beauty_GAN team. Those responsible gathered the most
popular and relevant beauty looks they could find, imagery as diverse and colourful as possible, with specs like ‘full face in shot’. They then sorted the imagery into categories and fed it into what is called a discriminator network, where the algorithm begins to learn stereotypical things about the images. It learns to distinguish an eye with make-up from an eye without make-up, or a smiling face versus a frowning face, for example. Eventually, the computer gets so good at distinguishing between categories that it is able to assign categories itself, to differentiate between a beauty selfie or, say, a picture of a dog.”

BeautyGAN provided a new aesthetic for the magazine and the beauty industry. What machine thinks of the beauty and the makeup is uncanny and weird. AI applied trendy makeup to Kylie Jenner, who started the trend, then it had an interesting look on it. When Kylie Jenner started a new trend, millions of people started to copy it. BeautyGAN took all the data and trained the model. When it got applied, it became different than what we thought. Apparently, the machine does not have the same idea as human beings, yet very interesting.