AI Smart Mirror
A digital AI smart mirror that connects with your wardrobe to help you choose the most suitable outfit by taking unexpected factors into account
62% of women experience frustration and stress when choosing an outfit. The average woman spends 17 minutes each day selecting an outfit, and men spend an average of 13 minutes daily deciding what to wear
My Role
Researcher UI/UX Designer
Target Users
Users that looks for confidence in what to wear
Year
March 2025
Teammate
Nuo Chen
How can we help users select the most suitable outfit to boost confidence and alleviate stress in choosing an outfit for events by considering unexpected factors such as weather, time, event type, mood, fashion trends, and the clothes they already own?
The increasing prevalence of mental health issues among older adults highlights the urgent need for innovative and accessible solutions to provide timely care and support.
Final Design
Final Prototype Mockup in Paper
I collaborated with my teammate to design and prototype the smart mirror. Once finalized, we printed it and attached it to a mirror for user testing.
Motivation
People who experience daily stress about what to wear, particularly before special events.
They often spend excessive time choosing outfits and struggle with closet organization, leading to frustration and decision fatigue.
Unforeseen situations and limited resources further heighten their stress and anxiety, underscoring the need for tools that help them adapt quickly and make confident decisions under pressure.
Recent Technologies
Conducted research on what are out there for closet and organzing clothes
Data Capturing (Smart Hanger)
-
The Near Field Communication (NFC)-enabled in-store display hanger is being used at a store in Paris.
-
It auto-detects which item is hanging on it.
-
Displays data based on inventory data on a built-in screen. For example, size, cost, and how many are in stock.
-
Use RFID tags, stores can capture both data for inventory management, as well as data for the consumer experience
AI-powered wardrobe organizer tools
-
Uses AI technology to simplify wardrobe management and provide personalized styling
-
Automatic background removal
-
Clothing attribute recognition
-
Outfit suggestions based on user preferences and existing wardrobe items.
-
The app’s AI stylist, LolaAI, also helps users make mindful shopping decisions by evaluating how new items fit with their current wardrobe.
Air Dresser
-
Sanitize and refresh clothes
-
It has a deodorizing filter that takes care of smells
-
An in-built Sanitize Cycle feature that produces high-temperature steam to remove 99% of common bacteria and dust
-
It can also be used as a dehumidifier. Just need to keep the door open so that your clothes smell like fresh linen.
-
Samsung AirDresser costs $1,400
Survey
Collected 31 responses. We want to understand people's beahvior and thoughts
Have you ever felt rushed or stressed when picking out clothes for an event or work?
45% yes
29% no
26% maybe
Would you want to be more efficient in this process?
48% yes
36% maybe
16% no
Would you be interested in using a phone or voice assistant to plan your outfits?
52% yes
29% no
19% maybe
If you’re gonna have a new smart closet, what features do you expect?
23 out of 31 people choose
AI recommendation for daily outfits
19 out of 31 people choose
Automatically record th clothes in the closet
12 out of 31 people choose
Reminder of washing/dry cloth
8 out of 31 people choose
Sharing outfits to friends/ Let friends choose
User Circumstance & Unexpected Situation
Here are some of the circumstance we think of based on our own experience and from what we researched
User Flow
Identified two key situations and developed the corresponding user flows.
Concept
Our ideas
-
Smart Mirror
When people stand in front of the “digital mirror”, it will automatically capture their current outfit and store it. Then, the mirror will use AI to generate outfit suggestions based on the clothes it has captured, the body size you entered, the style you prefer, the event you are attending, and the weather for the day. -
Rotating Rack
A closet where it will “give” people the clothes. Before opening the closet door, there is a screen on the closet that lets people see what clothes are inside by scanning the closet. People can pick the clothes they want to wear from the screen. After they know what clothes they want, they can select them. When they open the door, the clothes will turn into the ones they need, and they can just grab them without the need to dig into their closet.
Final Design Features
Final Prototype in detail
Camera Scanning
Final Prototype in detail
swipe to start or say "Scan my outfit", or any phrase using the word 'scan' to activate the camera to record the clothes
scans and tracks body changes over time
allowing users to better understand themselves and find the right clothing sizes
3 Suggestions optimized for comfort and functionality
Final Prototype in detail
Occasions
Color
Mood
User rating feature to support ongoing AI optimization
Final Prototype in detail
AI learning from the user's outfit selections while using the mirror
Dislike the suggestion
AI learn it and give a new suggestion based on user's choice
Virtual Try-on
Final Prototype in detail
Have the options to try outfit on virtually
AI will randomly give users three different backgrounds. (weather also change based on the background location)
User can choose their own background by saying the place name
Hand Gesture Navigation
Final Prototype in detail
Layout for different height
Final Prototype in detail
Another function of the camera is to intelligently adjust the mirror’s interface layout based on the user's height, ensuring it is positioned at an optimal level for ease of use. This enhances accessibility and provides a more user-friendly experience.
User Tests
To set up an environment for testing the Smart Mirror project, key features are selected and printed in a 1:1 ratio to ensure accurate evaluation.
We conducted 3 user testings
Female
senior college student
Female
senior college student
Female
Junior Designer
Part 1
By manually arranging and pasting the necessary components onto the mirror, new layouts can be quickly created for further assessment. This approach allows testers to observe real-time changes on each page, encouraging them to verbalize their thoughts while making decisions.
Part 2
Engaging testers with the Figma interface allows for a deeper understanding of the entire process. They will navigate the full interactive experience, exploring key functionalities and providing feedback on usability, design, and user experience. These high-fidelity interfaces offer enhanced details, realistic interactions, and additional features.
Findings
User Feedback
Positive
-
Love the AI-powered outfit selection feature, which helps reduce time spent on coordinating outfits and avoids unnecessary repeat purchases.
-
Also appreciate the 'Change Background' feature, as it creates an immersive experience that helps them visualize how the selected outfit would look in different settings.
-
The testing process comprises two approaches: a product test and an interface test, which together provide users with a more intuitive and immersive experience of the Smart Mirror.
Pain points
-
There are some uncertainties regarding interface details, as users sometimes struggle with what to do next and require guidance. For example, the purpose of the back button is not always clear.
-
One tester suggested enabling the ability to switch between tops and bottoms without having to click into the card, reducing an extra step for a more convenient experience.
Next Step
More features
-
Analyze the user's dressing style based on what have scanned
-
Share with friends
-
An overview of all clothes in the closet scanned by the mirror
-
Reminder of important events to have AI suggest outfit for particular event — link with calendar
-
AI generate outfits with clothes not in your closet — help people shop for new clothes
Different mirror shape