Face correlation with dlib
summary
A pipeline extracts faces from conference photos using dlib face recognition, converts each face to an N‑dimensional vector, and clusters results with Chinese Whispers to group likely same people. Small faces require an explicit scale image ×2 upsample to detect. A Qt front end displays and sorts images by face count or face area. Histogram correlation experiments fail, the face grouping works well, and a social‑graph view built from image–face links is implemented but not yet shown.
AI-written summary · about
related talks
01h 12mBuilding a live face recognition system
39mBuilding Highly Scalable Facial Recognition Pipelines
33mDo I Know You? Real time facial recognition with an Apache stack
38m3D Face Tracking and Reconstruction using Modern C++
30mThe Art and Science of Deep Face Recognition in Python
15mC-Face: Continuously reconstructing facial expressions by deep learning contours of the face ...
52mFace Recognition: From Scratch To Hatch
30mIn the service of the history. AI in archivistics.