The Technology Behind RecognizeMe AI

Understanding how our browser-based AI identifies and verifies identity in real-time.

01. Face Detection

When you start the tool, your browser uses a Convolutional Neural Network (CNN) to scan your webcam feed. It looks for specific patterns of light and shadow that represent a human face.

02. Landmark Mapping

Once a face is found, the AI identifies 68 unique landmarks. These include the contours of your jawline, the distance between your eyes, and the shape of your nose and mouth.

03. Feature Extraction

The AI converts these visual landmarks into a Mathematical Descriptor (a vector of 128 numbers). This "faceprint" is unique to you, much like a digital fingerprint.

04. Verification

The system compares your live descriptor against the one stored in your Firebase Firestore database. If the mathematical distance between them is small enough, the AI confirms a match.

Privacy-First Processing

Unlike other AI tools, RecognizeMe AI processes all data client-side. This means your video stream never leaves your computer. The AI calculations happen in your browser using your device's GPU/CPU power.