Skip to content

Tasfia-17/Inclusift

Repository files navigation

InclusiFit

AI-Powered Adaptive Fashion and Beauty for People with Disabilities

Built for the Perfect Corp x Startup World Cup Silicon Valley Hackathon 2026

InclusiFit Architecture


The Problem

1.3 billion people globally have disabilities. They shop online more than anyone because physical stores are inaccessible, exhausting, or humiliating. Yet every e-commerce platform is built for one body type.

Stat Source
71% abandon inaccessible e-commerce immediately Business Disability Forum
62% of disabled shoppers cannot find comfortable clothes RIDC 2024
$47.3B in adaptive clothing demand goes unmet in the US alone Coresight Research
$8 trillion global disability spending power Return on Disability

The gap is not just filters. It is visualization. Even if you find an adaptive product, you cannot see how it looks on your body with shorter limbs, wider feet, an AFO brace, or a hearing aid. You order blind. You return. You give up.

InclusiFit fixes this. Disability-aware smart filters combined with Perfect Corp's AI/AR virtual try-on creates the first platform where people with disabilities can find clothes that fit their life and see how they look before buying.


Solution

Two layers working together:

Layer 1 - Adaptive Discovery Engine

Smart filters that do not exist anywhere else:

  • Closure type: magnetic snap, velcro, zip, button-free
  • Waistband type: elastic, adjustable, rigid
  • Fit style: loose, relaxed, compression
  • Length: petite, standard, tall
  • Footwear: AFO-compatible, wide width, small adult sizes
  • Beauty: container type (pump, squeeze, twist), grip difficulty
  • Voice commands: "show loose fit", "AFO only", "clear filters"

Layer 2 - AI/AR Virtual Try-On (Perfect Corp YouCam API)

After filtering, users see themselves in the product before buying.


Perfect Corp APIs Integrated (13 total)

API Endpoint Adaptive Use Case
AI Clothes VTO v3 /s2s/v2.0/task/cloth Proportional fit, loose fit, closure visualization
AI Shoes VTO /s2s/v2.0/task/shoes AFO-compatible footwear, wide and small sizes
AI Skin Analysis HD /s2s/v2.1/task/skin-analysis Accessible beauty recommendations
AI Makeup VTO /s2s/v2.0/task/makeup-vto Try before buying for motor disability
AI Hairstyle Generator /s2s/v2.0/task/hair-style Wig and hairpiece try-on for alopecia
AI Hair Color VTO /s2s/v2.0/task/hair-color Color try-on for wig shoppers
AI Earring VTO /s2s/v2.0/task/2d-vto/earring Hearing aid compatible jewelry
AI Ring VTO /s2s/v2.0/task/2d-vto/ring Jewelry try-on
AI Bracelet VTO /s2s/v2.0/task/2d-vto/bracelet Adaptive clasp jewelry
AI Watch VTO /s2s/v2.0/task/2d-vto/watch Easy-clasp watches
AI Necklace VTO /s2s/v2.0/task/2d-vto/necklace Magnetic clasp necklaces
AI Bag VTO /s2s/v2.0/task/bag Adaptive bag try-on
AI Hat VTO /s2s/v2.0/task/hat Hat try-on
AI Text-to-Image /s2s/v2.0/task/text-to-image Generate adaptive fashion visuals
AI Face Analyzer /s2s/v2.0/task/face-attr-analysis Face attributes for personalization

The Five User Journeys

Journey 1: Shorter Limbs (Achondroplasia, Dwarfism)

  1. Profile: select "Shorter limbs"
  2. Catalog auto-filters: petite length, high waist rise
  3. Select dress, upload photo, AI Clothes VTO renders it on their body
  4. See actual length on actual body before buying

Journey 2: Limited Dexterity (Muscular Dystrophy, MS, Parkinson's)

  1. Profile: select "Limited dexterity"
  2. Fashion filters: velcro/magnetic closures, elastic waistband, loose fit
  3. Beauty filters: pump dispensers, easy-grip applicators
  4. AI Clothes VTO confirms loose fit, AI Makeup VTO shows color before application

Journey 3: AFO / Brace Users (GNE Myopathy, Cerebral Palsy)

  1. Profile: select "AFO / brace user"
  2. Catalog shows only AFO-compatible footwear
  3. Upload foot photo, AI Shoes VTO renders shoe on their foot

Journey 4: Hair Loss (Alopecia, Cancer Treatment)

  1. Profile: select "Hair loss / wigs"
  2. Upload selfie, AI Hairstyle Generator tries on wig styles
  3. AI Hair Color VTO shows color options
  4. Shop from home with privacy and confidence

Journey 5: Visual Impairment

  1. AI Skin Analysis results read aloud via Web Speech API
  2. Products recommended by concern and container type
  3. Voice commands: "show loose fit", "read my results", "AFO only"
  4. Screen reader compatible throughout

Pages

Page URL Description
Landing / Cinematic hero with glassmorphic panels
Onboarding /profile 3-step: welcome, interactive demo, condition selector
Dashboard /dashboard Quick actions, analytics, settings
Catalog /catalog 15 adaptive products, advanced filters, voice commands
AI Studio /studio Upload once, APIs run sequentially based on your profile
Clothes VTO /tryon Virtual try-on with before/after comparison
Shoes VTO /shoes AFO-compatible footwear try-on
Beauty /beauty Skin analysis and makeup VTO
Hair and Wigs /hair Hair color and wig style try-on
Accessories /accessories Ring, bracelet, watch, necklace, bag, hat VTO

Accessibility Features

Feature Who It Helps
Voice commands Motor disability, visual impairment
Skin results read aloud (Web Speech API) Visual impairment
Screen reader compatible (WCAG 2.1 AA) Visual impairment
Large tap targets (min 44x44px) Motor disability, tremors
Session save (localStorage) Fatigue conditions (MS, ME/CFS)
Before/after comparison All users
Free returns for users with disabilities All users
Caregiver mode toggle Severe motor disability
High contrast mode toggle Low vision

Adaptive Product Catalog (15 products)

  • Clothing (5): Adaptive Wrap Dress, Velcro Tunic, Pull-On Trousers, Magnetic Zip Jacket, Adaptive Midi Skirt
  • Footwear (3): AFO-Compatible Sneaker, Wide-Fit Loafer, Adaptive Boot with Side Zip
  • Beauty (3): Easy-Pump Foundation, Large-Grip Lip Color, Easy-Grip Mascara
  • Hair (2): Natural Wave Wig, Short Pixie Wig
  • Jewelry (2): Hearing Aid Friendly Studs, Magnetic Clasp Necklace

Each product is tagged with adaptive metadata: closure type, waistband, fit style, AFO compatibility, container type, grip difficulty.


Photon Spectrum Integration

InclusiFit extends beyond the web app to messaging platforms via Photon Spectrum. Users interact with InclusiFit directly in iMessage, WhatsApp, or Telegram without opening a browser.

How it works

The bot lives in the bot/ directory. It connects to Photon's hosted infrastructure and handles two types of input:

Photo input: User sends a selfie. The bot calls the Perfect Corp Skin Analysis API and returns scored results in the chat.

Text commands: User types a condition or category. The bot returns filtered adaptive product recommendations with direct links to the web app for virtual try-on.

Example conversation

User:  hi

Bot:   Welcome to InclusiFit — adaptive fashion for everyone.

       Commands:
         afo         — AFO-compatible footwear
         loose       — Loose / easy-wear clothing
         dexterity   — Magnetic/velcro closures
         hair        — Wigs and hair loss products
         beauty      — Accessible beauty products
         shorter     — Petite / shorter limb clothing
         catalog     — Full product list

       Send a photo of your face for a free AI skin analysis.
       Web app: https://inclusift.vercel.app


User:  [sends selfie]

Bot:   Analyzing your skin with Perfect Corp AI... (takes ~15 seconds)

       Skin Analysis Results:

         Pores: 72/100
         Moisture: 85/100
         Radiance: 68/100
         Oiliness: 74/100

       For personalized recommendations: https://inclusift.vercel.app/beauty


User:  afo

Bot:   AFO-Compatible Footwear:

       - AFO-Compatible Sneaker ($129.99)
         Extra wide, fits over AFO braces, velcro
       - Wide-Fit Slip-On Loafer ($79.99)
         Wide fit, slip-on, no laces

       Try them on: https://inclusift.vercel.app/shoes


User:  dexterity

Bot:   Adaptive Closures (Magnetic/Velcro):

       - Adaptive Wrap Dress ($89.99)
         Magnetic snaps, elastic waist, petite length
       - Easy-On Velcro Tunic ($49.99)
         Velcro closure, loose fit, front opening
       - Easy-Pump Foundation ($42.00)
         Pump dispenser, one-handed, SPF30

       Virtual try-on: https://inclusift.vercel.app/catalog?c=limited_dexterity

Bot setup

  1. Sign up at app.photon.codes to get your Project ID and Secret
  2. Navigate to the bot directory:
cd bot
cp .env.example .env
  1. Edit .env:
PROJECT_ID=your_photon_project_id
PROJECT_SECRET=your_photon_project_secret
PERFECT_CORP_API_KEY=your_perfect_corp_api_key
  1. Install and run:
npm install
node --env-file=.env index.js

Supported platforms

Platform Provider
iMessage spectrum-ts/providers/imessage
WhatsApp spectrum-ts/providers/whatsapp
Telegram spectrum-ts/providers/telegram
Terminal (testing) spectrum-ts/providers/terminal

To switch platforms, replace imessage.config() with the desired provider in bot/index.js.


Tech Stack

Layer Technology
Web framework Next.js 16 (App Router)
Language TypeScript
Styling Tailwind CSS + CSS custom properties
Font Inter (Google Fonts)
AI/AR Perfect Corp YouCam API (13 endpoints)
Messaging Photon Spectrum (iMessage, WhatsApp, Telegram)
Voice Web Speech API (built-in browser)
State React useState + localStorage
Deploy Vercel

Web App Setup

1. Clone and install

git clone https://github.com/Tasfia-17/Inclusift.git
cd Inclusift
pnpm install

2. Run

pnpm dev

Open http://localhost:3000

3. Deploy to Vercel

Import the repo at vercel.com. No environment variables required — the API key is bundled for the hackathon demo.


API Architecture

All Perfect Corp API calls go through Next.js Route Handlers (server-side). The API key is never exposed to the client.

Browser  ->  /api/vto/upload          ->  Perfect Corp file upload (pre-signed URL)
Browser  ->  /api/vto/clothes         ->  POST /s2s/v2.0/task/cloth
Browser  ->  /api/vto/clothes/[id]    ->  GET poll until task_status: success

Verified request formats:

Clothes VTO:

{
  "src_file_id": "...",
  "ref_file_url": "https://...",
  "garment_category": "auto"
}

Shoes VTO:

{
  "src_file_id": "...",
  "ref_file_url": "https://...",
  "gender": "female"
}

Skin Analysis:

{
  "src_file_id": "...",
  "dst_actions": ["hd_acne", "hd_pore", "hd_moisture"],
  "miniserver_args": { "enable_mask_overlay": true }
}

File Upload:

{
  "files": [{
    "file_name": "photo.jpg",
    "file_size": 123456,
    "content_type": "image/jpeg"
  }]
}

Business Case

Metric Value
Global disability spending power $8 trillion
US adaptive clothing market $47.3B untapped
AFO footwear market $251M growing to $390M by 2032
Higher conversion with AR vs static 94%
Fewer returns with VTO 64%
Benefit Cosmetics sales uplift from skin analysis 14x
Avon conversion boost with VTO 320%

Brand partnership potential: Tommy Hilfiger Adaptive, Nike FlyEase, Zappos Adaptive, Rare Beauty, Guide Beauty. All have adaptive product lines with zero VTO. InclusiFit is the missing layer.


Why This Wins

  1. Unexpected use case: no one else is building VTO for disability
  2. Real documented pain points: real conditions, real barriers, real community
  3. 13 Perfect Corp APIs: more than any competitor, each solving a specific problem
  4. $8T market: the biggest underserved retail segment on earth
  5. Social impact combined with business ROI: a rare combination
  6. Photon Spectrum: extends InclusiFit to iMessage and WhatsApp, meeting users where they already are
  7. Perfect Corp's own gap: their accessibility statement admits no speech output. InclusiFit fills it using their own APIs

One-Paragraph Pitch

InclusiFit is an AI-powered adaptive fashion and beauty platform that combines disability-aware smart filters with Perfect Corp's virtual try-on technology. People with disabilities, whether they have shorter limbs, limited dexterity, use an AFO brace, experience hair loss, or have visual impairment, can filter products by the features that matter to their body and their life, then see exactly how those products look on them before buying. The platform runs on the web, in iMessage and WhatsApp via Photon Spectrum, and as a Replit prototype. No more ordering blind. No more returning five items to keep one. We serve 1.3 billion people, $8 trillion in spending power, and a $47.3B adaptive market that every mainstream platform has ignored. Fashion should fit everyone.


License

MIT. Built for the Perfect Corp x Startup World Cup Hackathon 2026.

About

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors