Analyze, track, simulate
Skin Health
Mirra turns one quality-gated selfie into fourteen skin scores, skin age, tone context, trends, treatment simulation, and product direction.
14 concernsSkin ageBefore and after
Built for the Perfect Corp AI hackathon
Mirra
One good selfie powers your whole look—skin scores, face-aware makeup and hair, closet matching, and try-ons you can review before you buy.
Instead of juggling separate beauty apps, Mirra connects Perfect Corp analysis, weather and context, your wardrobe, and product search so you see what changed, what to try next, and why it fits—then decide with proof, not guesswork.
- 9
- AI APIs
- 1
- Selfie
- 90s
- Demo flow
What Mirra Shows
Everything about how you look, wired together—not a one-off try-on toy.
People do not want another disconnected score or random product grid. Mirra turns analysis into a path: what the scan means, what to simulate, how your closet covers the moment, what to try on, and what is worth buying.
Face-aware styling
GlowUp Studio
Face shape, undertone, hair, lip, eye, and brow colors become coordinated makeup, hair, earring, and necklace choices.
Tone-awareFace shapeVTO results
From wardrobe to proof
Closet and Try-On
Mirra ranks what you own, finds what you are missing, lets you try products on, and produces a proof card before purchase.
Closet matchGap searchProof card
Guided flow
See the “why,” then do the next step.
Mirra is built like an operator for your appearance: one capture, then analysis, context, previews, and a clear decision—not a black box that only shows a final image.
01
Capture once
A guided selfie flow checks lighting and framing so skin, face, and try-on features get a usable portrait—no mystery uploads.
02
Read the signal
Perfect Corp powers skin, tone, face attributes, and virtual try-on. Mirra folds those results into one profile you can actually use.
03
Add real-world context
Gemini ties together your scan, weather, history, closet, and product search into a visible reasoning trace—so recommendations match the day, not a generic model.
04
Decide with proof
Simulate skin improvements, preview makeup and hair, match outfits, and see a proof-style summary before you spend.
Skin signal
Moisture dropped while oiliness stayed stable.
Context
Low humidity increases dehydration risk today.
Action
Simulate hydration improvement and compare HA serum options.
API Depth
Nine Perfect Corp APIs become one consumer journey.
In one journey you can go from skin analysis to simulation, GlowUp-style makeup and hair, clothes and accessory try-on, and a proof-oriented summary—without hopping between disconnected demos.
01
Skin analysis
02
Skin tone
03
Face attributes
04
Skin simulation
05
Clothes VTO
06
Makeup VTO
07
Hair transfer
08
Earrings VTO
09
Necklace VTO
Proof Card
Buy after the look makes sense.
Tone match
97%
Style fit
94%
Skin-safe
Clear
New spend
$127
Retail Value
Confidence before checkout.
Before checkout, Mirra surfaces tone match, fit with your closet, skin compatibility, and what new spend would add—so try-on feels accountable, not like a filter.
Preview products on the user, not on a model.
Search real products when the closet has a gap.
Show why a recommendation fits before asking for payment.
Privacy and Control
Built around consent, session auth, and clear data use.
Google sign-in starts the flow, selfies are used to create the user profile, and product recommendations are generated from the user's own scan signals.
Auth-gated app
Context-aware tips
Transparent reasoning
User-first previews