POW- Personalized Wardrobe Assistant

Digitize Your Closet. Own Your Style.

Comment

Inspiration

The average consumer owns thousands of dollars in clothing, yet 80% of our wardrobes are "digitally invisible". I was inspired by the massive waste in the fashion industry—where sizing uncertainty drives return rates over 30%—and the simple frustration of staring at a full closet and feeling like I have nothing to wear. I wanted to build a "Personal Fashion Playground" that turns your phone into a magic mirror, allowing you to discover your style without the physical hassle.

What it does

  • Creates a Personalized 3D Avatar: Users can design a digital twin that mirrors their specific body shape and style, allowing for realistic visualization of fashion choices.
  • Digitizes Physical Wardrobes: The app automatically organizes your existing clothing inventory by removing backgrounds and using AI to recognize attributes like category and color from uploaded photos.
  • Enables Virtual Try-On: You can visualize how garments look and fit on your 3D avatar, helping you evaluate styles and colors before wearing or buying them.
  • Powers Interactive Styling: Users can mix and match pieces from their digital collection to create new outfits and explore endless combinations with a virtual styling companion.
  • Facilitates Private Social Sharing: The platform allows you to connect with friends, share your wardrobe privately, and collaborate on styling ideas in a secure environment.
  • Manages P2P Lending Requests: A formal workflow enables users to request and approve the lending of physical items within their trusted social circles.
  • Synchronizes with RFID Technology: In retail environments, the app can detect physical items you pick up and instantly project them onto your digital avatar.
  • Provides Smart Analytics: The app calculates insights like cost-per-wear to help you understand the true value of your items and make more mindful fashion decisions.

How we built it

Building POW (Personalized Outfit Wardrobe) was an exercise in "phygital" collaboration—the merging of physical textiles with digital twins. The Brain (AI): I leveraged Google Gemini's vision capabilities to automatically categorize and map garments from simple photos. The Body (3D Avatars): Instead of rigid templates, I used AI-driven 3D reconstruction to create digital twins that capture unique body shapes and proportions.

As a non-coder, I learned that I didn't need to master syntax to be a builder; I needed to master intent. By using AI Studio's Annotation Mode, I realized I could iterate on the UI just by highlighting components and describing changes in plain English. I learned to treat the AI as a collaborator, refining my app one prompt at a time to fix logic errors rather than trying to build a perfect system from the start.

Challenges we ran into

Moving from an idea to a functional web prototype in AI Studio presented specific hurdles:

The Web-Native Gap: While AI Studio writes valid mobile code (like Kotlin or Swift), its Live Preview is currently limited to web frameworks like React or Angular, meaning I had to rely on a web-based "vibe" before seeing the final iOS layout.

Logic vs. Prompting: Some complex tasks, like real-time 3D physics for fabric drape, required multiple iterations to ensure the AI's generated code didn't break during the "I'm Feeling Lucky" or automated generation stages.

Usage Limits: As a free-tier user, navigating daily rate limits meant I had to be extremely strategic with my prompts to avoid hitting the "Usage Limit" before finishing a key feature.

Accomplishments that we're proud of

Zero-Code MVP: Built a functional web-based wardrobe assistant without writing a single line of manual code.

AI Vision Integration: Successfully implemented Gemini-powered image recognition to categorize clothes automatically.

Identity-First Design: Created a prototype that places the user's 3D digital twin at the center of the experience.

What we learned

Founders can build: You don't need a CS degree to build a functional prototype; you need a problem to solve and a tool like AI Studio.

Prompting is the new coding: The quality of the app is directly tied to the clarity of the founder's instructions.

What's next for POW- Personalized Wardrobe Assistant

Native iOS Launch: Moving from a web prototype to a native app using Xcode and SwiftUI.

RFID Real-World Sync: Finalizing the hardware integration to link physical closet tags to the digital app.

P2P Lending Beta: Opening the private social network for our first 100 beta testers to share and lend clothes.

Built With

Share this project:

Updates