L'Oreal (marques ModiFace)
AR virtual try-on (synthetic makeup rendering on the face)
L'Oreal's ModiFace virtual try-on totaled more than 100 million sessions in 2023 (versus 40 million in 2022) and triples the conversion rate where it is deployed.
Key points
- AR virtual try-on of makeup on the face, online and in store.
- ModiFace engine (vision, deep learning, real-time rendering), L'Oreal subsidiary.
- More than 100 million sessions in 2023 (40M in 2022), conversion tripled.
- Evidence level A, status confirmed: figure stated in 2024 annual results.
Objective
Reduce purchase doubt on makeup (color, shade) by letting customers test products on their own face, and increase conversion and engagement on e-commerce and in store.
The deployment
ModiFace, acquired by L'Oreal in 2018, provides virtual try-on and beauty diagnosis to a large share of the group's brands (Maybelline, L'Oreal Paris, Lancome, Urban Decay, etc.) as well as to retailers and platforms (Amazon, A.S. Watson, DM). The user opens the camera, the system detects and tracks the face in real time, then applies a photorealistic rendering of the product (lipstick, eyeshadow, foundation). The same engine also powers skin diagnosis. Global deployment, on brand websites, at retailers, and through in-store mirrors/kiosks.
The case in action
Official videoL'Oreal Paris Virtual Try On (ModiFace) · voir sur YouTube
Results Proof A
The 100 million sessions figure stated by CEO Nicolas Hieronimus in an earnings presentation (February 2024), corroborated by the Chief Digital Officer (The Drum) and by the documented multi-retailer deployment (WWD).
How it works
Documented architectureThe stack in detail
- plateforme ModiFace virtual try-on and beauty diagnosis engine acquired by L'Oreal in 2018, used by the group's brands and by retailers (Amazon, A.S. Watson, DM)
- outil Moteur de detection et suivi du visage ModiFace proprietary deep learning: feature detection and real-time tracking before applying the photorealistic rendering
- infra SDK d'integration (web, app, borne) component embedded on product pages, apps, and in-store mirrors/kiosks; inference most often runs locally on the device
- infra Referentiel de teintes calibrees product/rendering mapping (shades, textures) maintained by the teams for each collection; this is what makes the try-on reliable
How it runs, concretely
For ops teams-
1Opening the try-on client
The user activates the camera on the product page or the in-store kiosk.
-
2Face detection and tracking AI (ModiFace)
The engine detects the features and tracks the movements in real time.
-
3Product rendering AI (ModiFace)
The selected shade/texture is applied photorealistically and follows the face like a mirror.
-
4Comparison and add to cart client
The user goes through looks, compares before/after, and moves to purchase.
-
5Calibration and catalog brand / ModiFace team
The teams maintain the product/rendering match with each new collection.
The precise mapping of product references (shades, textures) onto the rendering. If a shade is not correctly calibrated, the try-on misleads and breaks purchase confidence.
How your customers perceive this type of use
Sourced studiesUn ecart net separe les annonceurs des consommateurs : 77% des annonceurs voient l'IA positivement contre 38% des consommateurs (Yahoo/Publicis, 2024). Les mesures implicites confirment le rejet declare : en EEG, les pubs generees par IA produisent une activation memorielle plus faible que les pubs traditionnelles et sont decrites comme agacantes, ennuyeuses et confuses (NIQ, 2024). La disclosure a un effet ambivalent : elle augmente fortement la confiance quand elle est remarquee (Yahoo/Publicis), mais 27% des jeunes consommateurs disent faire moins confiance a une entreprise dont la pub est creee par IA (IAB, 2024).
Acceptance conditions
- Une disclosure visible : quand la mention IA est remarquee, la confiance globale envers l'entreprise augmente de 96% (Yahoo/Publicis 2024)
- Une qualite visuelle suffisante : les visuels IA de basse qualite augmentent l'effort cognitif et distraient du message (NIQ 2024)
Red lines
- Le contenu IA non declare puis identifie : 72% des consommateurs disent que l'IA rend l'authenticite difficile a etablir (Yahoo/Publicis 2024) et les marques utilisant des pubs IA sont plus souvent jugees inauthentiques ou non ethiques par les consommateurs que par les dirigeants (IAB 2024)
- Les mannequins et personnes generes par IA : 46% des consommateurs n'en veulent pas dans la publicite, l'inquietude premiere etant les standards de beaute irrealistes (Attest 2025)
Sources: Yahoo / Publicis Media (terrain Ebco) 2024 · IAB (avec Attest) 2024 · NIQ (NielsenIQ) 2024 · Attest 2025
How to replicate
Inference, not sourcedData prerequisites
- product reference with shades/textures calibrated for the AR rendering
- product visual assets
Org prerequisites
- e-commerce integration (product page, app)
- calibration process for each collection
Possible stack
- ModiFace
- Perfect Corp (YouCam)
- Banuba
- Wanna/AR SaaS
The plan, step by step
- Step 1Choose the highest-traffic makeup product pages and contract a try-on SaaSDeliverable: List of pilot SKUs + signed SaaS contract
- Step 2Calibrate the priority shades and textures with the vendorDeliverable: Shade/rendering reference visually validated by the business
- Step 3Integrate the SDK on the pilot pages with camera consent and no image retentionDeliverable: Try-on in production on the pilot pages, GDPR-compliant
- Step 4Measure the effect on conversion and engagement in A/B (with vs without try-on)Deliverable: Quantified read of the conversion lift and time spent
- Step 5Extend to the rest of the catalog, to the apps, and to in-store kiosks, with calibration for each collectionDeliverable: Rollout plan by collection + recurring calibration process
First step: Connect a try-on SaaS to the highest-traffic makeup product pages and calibrate the priority shades.
The same pattern, proven in other industries
Leroy Merlin (Adeo)
genAI enrichment of product attributes and pages
eBay
GenAI generation of a product listing from a photo
Intuit
done-for-you genAI assistant integrated into the product to reduce friction and convert
Sources
- S1 L'Oreal Sees 150% Increase in Virtual Try-Ons as Consumers Seek AR Immersion (cite l'earnings call, Nicolas Hieronimus) Secondary archive pending
- S2 Conversion rates triple when L'Oreal uses AR tech to showcase products Established press archive pending
- S3 Amazon Implements L'Oreal's ModiFace Virtual Makeup Try-ons Established press archive pending
An error, newer info, a source?
This page lives on its accuracy. If a figure has moved, if the deployment has changed, or if you have a higher-quality source, tell us. Every sourced correction is verified before publication.