About Golden Face Ratio

I'm a computer science researcher specializing in computer vision and facial landmark analysis. I've spent the past two years studying how convolutional neural networks detect and interpret facial geometry — and more importantly, what those measurements actually mean (and don't mean) in the real world.

Why I Built This Tool

The idea started when I watched a teenager get upset over a 61% score from one of those face analysis apps. She treated it like a grade. It wasn't — it was a geometric measurement with zero context. But the app didn't explain that. Most of them don't.

I wanted to build something different. A tool that runs entirely in your browser, explains what each measurement actually means, and doesn't store a single pixel of your data. Most existing tools were either locked behind paywalls, required server-side photo uploads, or gave vague results without showing the underlying math.

Golden Face Ratio is my attempt to bridge that gap — accurate computer vision technology paired with honest, plain-language explanations. The measurements are real. The context matters more.

How It Works: The Tech Stack

The site is built with Next.js for server-rendered pages and SEO. The facial analysis engine uses Google's MediaPipe Face Mesh — the same 468-landmark detection model used in published academic research on facial geometry (see Kartynnik et al., 2019). All processing happens client-side via WebAssembly and WebGL, so your photos never leave your device.

When you upload a photo, the AI identifies 468 landmark points, calculates pixel distances between key feature pairs, and compares the resulting ratios to φ = 1.618. You get five individual proportion measurements and an overall harmony score — all in under five seconds. The methodology is explained in detail in the calculator guide.

Research Methodology

The five facial proportions we measure are drawn from peer-reviewed studies on facial attractiveness and proportional analysis. Key references include:

• Pallett, Link & Lee (2010) — established updated "golden" ratios for facial beauty through controlled perception experiments (Vision Research, 50(2))
• Bashour (2006) — comprehensive review of facial attractiveness analysis methods in clinical settings (Plastic and Reconstructive Surgery, 118(3))
• Schmid, Marx & Samal (2008) — computational approaches to measuring facial attractiveness indices (Pattern Recognition, 41(3))

My Research Focus

• Facial Landmark Detection and Geometric Analysis
• Privacy-Preserving AI (Client-Side Processing via WebAssembly)
• Algorithmic Bias in Beauty Standard Modeling
• Cross-Cultural Variation in Perceived Facial Harmony
• Accessible Computer Vision Tools for Non-Technical Users