You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
"description": "A trillions of old photos are a valuable source of information about the\npast. Unfortunately most of them are not described sufficiently or at\nall. Imagine Artificial Intelligence as a friend of the archivist of the\n21st century. This is the end of unlabeled photos epoch.\n\n1839 is a data generally accepted as the birth year of practical\nphotography. Since then mankind produced about 10 quadrillions of photos\nincluding 1 quadrillion only last year. This huge amount of unlabeled an\nundescribed data is a problem, if we want to obtain important\ninformation quickly and efficiently. Old photos are extremely valuable,\nbecause they contain a lot of data about the past. However some\nexpertise and experience is needed to properly describe such images.\nWhat if we include all this knowledge into neural networks? Can AI\nbecome a friend of the 21st century archivist? Let\u2019s talk about\nautomatic image tagging and faces recognition in old photos.\n",