Online pic searches now automated
New image recognition software has been developed, which can identify online pictures.
An international team has revealed an image can be identified by 256 to 1,024 bits of data.
Led by Antonio Torralba, an assistant professor at the Massachusetts Institute of Technology, the researchers hope the breakthrough "could lead to great advances in the automated identification of online images and, ultimately, provide a basis for computers to see like humans do".
They managed to store about 13 million images from the internet in a database of only 600 MB and claimed to be able to retrieve one in less than a second.
Their results will be published as an academic paper, Small Codes and Large Image Databases for Recognition.
Mr Torralba and colleagues hope their findings will develop efficient image search and "scene matching techniques that are not only fast, but also require very little memory, enabling their use on standard hardware or even on handheld devices".
According to Analysts, Nielsen Online, video social networking is the fastest online social trend – top site YouTube.com attracted more than ten million viewers last year.
