It seems that the world of artificial intelligence is set to gain another leap forward, where the brilliant minds over at Google have come up with a new generation of artificial intelligence software which will be able to describe the contents of photographs in a manner that is a whole lot more accurate than ever before. In fact, the artificial intelligence’s aim is to deliver a description for the image it “sees” in a manner that is not too different from that of a human mind.
In the long run, this would make it a whole lot easier to look for images, as the software can then be used to assist the blind and visually impaired to understand pictures in a more comprehensive manner. Google are not the only ones looking into this possible future, as Stanford University too, have experienced a breakthrough in the similar field. The machine-learning software that Google worked on will make use of a couple of neural networks, where one of them deals with image recognition, while the other would rely on natural language processing.
The quartet of scientists behind this software would be Oriol Vinyals, Alexander Toshev, Samy Bengio and Dumitru Erhan, and it truly brings to life the proverb “A picture may be worth a thousands words.” [Press Release]