>_ TECHNOLOGY
How Face ID Search Works
Our reverse face search engine uses neural network-based facial embeddings to find matches across 200+ platforms and 50M+ indexed faces.
$ PROCESS
$ upload --file photo.jpg
Face detected. Extracting 128 unique facial data points. Generating vector embedding for comparison.
Upload any clear photo containing a face. We support JPG, PNG, and WEBP formats. The image is processed locally before being sent to our secure servers.
$ scan --deep --platforms all
Scanning social media, dating sites, news archives... Cross-referencing 50M+ indexed face vectors.
Our AI engine converts the face into a 128-dimensional vector and compares it against our index of 50M+ faces from social media, dating platforms, news sites, and public records.
$ results --format detailed
12 matches found. Generating report with source links, similarity scores, and platform metadata.
Get detailed results with direct links to matching profiles, similarity scores, and platform information. Results typically arrive in under 60 seconds.
$ CAPABILITIES
Vector Embeddings
128-dimensional facial vectors enable sub-second matching against massive datasets with high accuracy even across varying conditions.
Deep Web Indexing
Our crawler indexes faces from 200+ platforms including social media, dating sites, news archives, and public databases.
Privacy First
Uploaded images are deleted immediately after processing. No facial data is stored. All searches are anonymous.
Real-time Processing
Results in under 60 seconds using distributed GPU infrastructure for parallel vector comparison.
Multi-angle Detection
Works across different angles, lighting conditions, and age variations using robust facial landmark detection.
OSINT Ready
Built for investigators, journalists, and security professionals. Comprehensive reports with direct source links.