Researchers at the Massachusetts Institute of Technology (MIT) have designed a machine-learning system that appears to have worked out itself a key factor in how the human brain recognises people’s faces. The team trained the system to recognise faces by feeding it a set of sample images. They found an intermediate processing step had emerged, which took into account the angle of rotation of the face being scanned, spontaneously duplicating a feature seen in primates.

Next: Visit the NFCW Expo to find new suppliers and solutions