The algorithm detects faces regardless of lighting, expression or whether or not the individual is facing the camera.
Facebook has developed a new algorithm called DeepFace, which it claims can recognise human faces with 97.25% accuracy.
The social networking site said the new software could also detect faces regardless of lighting, expression or whether or not the individual is facing the camera.
It has reportedly reduced the error rate by over a quarter compared to similar software earlier on, marking a major advancement over earlier face-matching technologies.
Facebook AI team member Yaniv Taigman said you normally don’t see that sort of improvement.
"We closely approach human performance," he said.
DeepFace is technically a facial verification system, which matches images to images, not facial recognition software, which matches images to a name.
The DeepFace algorithm can identify up to 4,000 identities based on a database of over 4 million separate images in its current version.