Google Inc. unveiled a new addressable unable to identify the content of images, videos and automatically give the titles and descriptions of the appropriate use of artificial intelligence and natural language technology system.
However there are many programs, so-called "computer vision" able to add titles to the images, but Google's new system, characterized according to the company, with a high degree of accuracy makes it difficult to distinguish between them and the titles set by humans.
Experts believe that, despite the fact that the research project is still in its early stages, but it heralds a promising future in the field of artificial intelligence and machine learning techniques.
Google said on her blog, "This type of system can help in the end people visually impaired to understand the images, and to provide an alternative for images text in parts of the world where communications are slow phones, also makes it easier for everyone search on Google for images."
Post a Comment