Overview
ποΈ Getting Started
Set up token, permissions, first scan
ποΈ System Requirements
Devices, camera specs, native vs. web
ποΈ SDKs
Android, Web, React Native integration
ποΈ API Reference
Endpoint structure, parameters, payload
GlamARβs AI Facial Skin Analysis uses deep learning and computer vision to evaluate skin health in real-time. With just one camera capture, it can detect 14+ skin conditions β from acne and wrinkles to pigmentation and pores β and return visualized, annotated results.
No wearables or external hardware required.
What Is It For?β
- Provide personalized skincare insights to customers
- Power AI-driven product recommendations
- Enable skin assessments in apps, kiosks, or smart mirrors
- Support dermatology-lite solutions for beauty, wellness, and medtech
What It Detectsβ
Parameter | Category | Type | Range / Values | Annotated Image | Notes |
---|---|---|---|---|---|
Skin Score | Overall | Integer | 0β100 | β | Overall score based on all sub-metrics |
Skin Type | Classification | Class | oily, dry, normal, combination | β | Based on U-zone and T-zone variation |
Skin Tone | Classification | Class | fair, light, medium, olive, tan, deep | β | Melanin-based classification |
Skin Age | Classification | Integer | 18β80 | β | Perceived age in years |
Acne | Key Condition | Integer | 0β100 | β | Whiteheads and inflamed pimples detection |
Wrinkles | Key Condition | Integer | 0β100 | β | Fine lines and deep fold estimation |
Pores | Key Condition | Integer | 0β100 | β | Visibility and density of pores |
Eye Bags | Eye Area | Integer | 0β100 | β | Lower eyelid puffiness |
Under Eye | Eye Area | Integer | 0β100 | β | Orbital fat changes |
Dark Circles | Eye Area | Integer | 0β100 | β | Hyperpigmentation under the eyes |
Scars | Specific Issue | Integer | 0β100 | β | Depth and presence of scars |
Whiteheads | Specific Issue | Integer | 0β100 | β | Closed comedones evaluation |
Pigmentation | Specific Issue | Integer | 0β100 | β | Local melanin concentration |
Hydration | Specific Issue | Integer | 0β100 | β | In development |
Redness | Specific Issue | Integer | 0β100 | β | In development |
Projected Skin (1) | Projection | Integer | N/A | β | Filtered forecast: 1-month outcome |
Projected Skin (2) | Projection | Integer | N/A | β | Filtered forecast: 3-month outcome |
Integration Optionsβ
Method | Use Case |
---|---|
SDKs | For real-time camera capture and instant analysis (Web, Android, iOS) |
APIs | For backend-based analysis using image input |
Plugins | Shopify, Webflow, Mirror UI modules |
Console | Upload datasets, test reports, view logs |
π¦ SDK is ideal for real-time camera mode.
π‘ API is best for batch or backend-triggered scans.
How It Works (Under the Hood)β
- Image Capture β via live camera or uploaded image
- ML Inference β Face is analyzed using trained CNN models
- Result Generation β 14+ skin metrics + annotated photo
- Visual Feedback β Returned to client as JSON with image
Models are optimized to run in under 3 seconds on-device or .5s via cloud.