Back to timeline

Research · June 2014

DeepFace: face recognition near human level

In June 2014 Facebook AI Research and Tel Aviv University presented DeepFace at CVPR: a network of over 120 million parameters, after a three-dimensional alignment of the face, reached 97.35% on Labeled Faces in the Wild, cutting the error by over 27%.

Why it matters

Face verification in unconstrained photographs came close to human level, which the paper itself gives as over 97.5%. Its introduction says outright that the machines' lag had so far been a buffer against the social and cultural implications of face recognition.

Trained on the SFC set: 4.4 million faces of 4,030 people, 800 to 1,200 each. The face is aligned by a three-dimensional model and 67 fiducial points. On YouTube Faces the error fell by over 50%. The human level on LFW is taken from other work, not measured.

Event record

Event date
June 2014
Timeline date
Event date
Verification
Sources gathered automatically · September 25, 2026
Lines
ID
evt-0755

The CVPR 2014 proceedings by Crossref (June 2014); the conference ran 23-28 June in Columbus. The day the proceedings appeared was not read.

Sources

Related events

Records that link to this one