Face Metrics 2.0
Purpose, Scope and Limitations
Disclaimer
Face Metrics 2.0 Application Algorithms are not intended to be a substitute for professional medical advice, diagnosis or treatment.
None of the provided skin health parameters estimations can be treated as health advice.
Always seek the advice of your physician or other qualified healthcare provider on any medical condition or treatment as well as before undertaking any healthcare regimen.
Face Metrics 2.0 Application provides Algorithms that perform the analysis of skin visual appearance on the facial area. The analysis is aimed at evaluating the visible and measurable skin condition parameters.
Algorithms included in this Application
Skin Health Algorithms
These Algorithms are designed to evaluate visible and measurable skin health parameters:
Acne algorithm, see more here: Acne Algorithm.
Eye Age algorithm, see more here: Eye Age Algorithm.
Eye Area Condition algorithm, see more here: Eye Area Condition Algorithm.
Hydration algorithm (experimental), see more here: Hydration Algorithm (experimental).
Lines algorithm, see more here: Lines Algorithm.
Perceived Age algorithm, see more here: Perceived Age Algorithm.
Pigmentation algorithm, see more here: Pigmentation Algorithm.
Pores algorithm, see more here: Pores Algorithm.
Redness algorithm, see more here: Redness Algorithm.
Skin Tone algorithm, see more here: Skin tone Algorithm.
Skin Type algorithm (new since Aug 2024), see more here: Skin type Algorithm
Uniformness algorithm (experimental), see more here: Uniformness Algorithm (experimental).
Several algorithms are available for a limited set of clients, but better alternatives are recommended for everyone:
Translucency algorithm -> Instead, use the Skin Type algorithm.
ITA (Individual Typology Angle) algorithm -> Instead, use the Skin Tone algorithm.
Wrinkles (deprecated) algorithm -> Instead, use the Lines Algorithm.
Sagging (deprecated) algorithm -> Instead, use the Lines Algorithm.
Technical Algorithms
These Algorithms are designed to provide additional technical capabilities for visualization of analysis results.
Quality, see more here: Quality Algorithm
Facial Landmarks
Availability of Application
Face Metrics 2.0 Application is used to analyse all images, that are collected via Skin Consultant App for all plan clients.
Face Metrics 2.0 Application is available via API for Professional plan clients. See the detailed API doc: API for Face Metrics 2.0.
Input Requirements
Input Data
Face Metrics 2.0 Application requires a high-resolution portrait photograph of a human face with:
Open eyes and neutral facial expression.
No face-covering accessories or obstructions.
Clear skin, no make-up.
Face Metrics 2.0 Application supports input in the form of:
A single frontal portrait image.
A triplet (3) of frontal, right 40-degree, and left 40-degree portrait images.
Dependency on Other Applications
The Face Metrics 2.0 Application does not depend on other applications and can be used directly with input data.
Output Details
Output Data
Face Metrics 2.0 Application returns a variable set of numerical and categorical estimations of visible and measurable skin health parameters, evaluated with Algorithms included in this Application.
For the Advanced+ plan, the output data is available in Subjects photos interface.
For the Professional plan, the output is available via API, see Generic Algorithm Metric Scheme.
Dependency for Other Applications
Face Metrics 2.0 Application could be used as an input data in order for the following applications to be executed:
FAQ
What is the actual processing time for FaceMetrics 2.0?
For Face Skin Metrics 2.0, the processing time averages under 5 seconds for 95% of images.
How accurate is FaceMetrics 2.0?
FaceMetrics 2.0 algorithms were trained on over 30 million real skin images. The algorithms are validated by dermatologists, ensuring high accuracy across diverse skin types and ethnicities. Some clients have reported up to 50% increase in skin condition prediction accuracy after switching to our platform.
For more details, please reach out to your customer success manager.
For details about each algorithm's potential variability, please check the disclaimers on each algorithm page.
Last updated
Was this helpful?