Now scientists have developed a computer system which can help to discover and identify the real world objects which are based on the same method of visual learning that humans nowadays mainly use the system is an advance in a type of technology known as the “computer vision” that also nables the computers to read and identify the visual images..
It could be going to be an important step towards the general Artificial Intelligence system computers that learn on their own are intuitive, that even makes the decision based on reasoning and interact with the humans in a much more human-like way.
Moreover nowadays current AI computer vision systems are increasingly powerful and capable, they are meaningful and task-specific that their ability to identify what they see is also limited by how much they have been programmed and trained by the humans.
Humans can also easily intuit where the dog’s head and the rest of its body are, but that ability will also elude with most of the Artificial Intelligence system, some of the researches said. Some of the current computer vision systems is not just designed to learn on their own.
The approach is made up of three broad steps. First, the system breaks up an image into the small chunks, with the researchers call “viewlets”, secondly, the computer learns how those viewlets fit together to form the object in question.
“One is a wealth of images and videos that depict the same types of objects. The second is that those objects are shown from many perspectives-obscured, bird’s eye, up-close-and they are placed in all different kinds of environments,” Roychowdhury said.