Researchers from Purdue University in the United States are trying to make smartphones capable of recognizing and understanding objects that are in the visual field of the integrated camera.
The research focuses on the capability called deep learning, which could be loosely translated as deep learning, as different learning layers are used that mimic the way information is processed by the human brain.
The same technique is used in software that allows search engines to locate images and video files that have been tagged with certain keywords. However, that deep learning algorithm cannot be directly transferred to portable devices, as they lack the necessary processing power.
Thus, the researchers proceeded with the necessary modifications in order to develop the software that would allow them to use the algorithm from the processor of a conventional smartphone. The new deep learning capability that was developed gives portable devices an upgrade that reaches the limits of artificial intelligence, according to the researchers.
The new, customized deep learning technique for mobile devices also works at «sequential» levels. Researchers at Purdue University consider that the full development of this technology is essentially a matter of time and it may not be long before it is integrated into commercially available smartphones and wearables.
The research results could be used in applications such as Google Glass, facial recognition systems, or even in robotic autonomous vehicles.
The research program is the result of the collaboration of the U.S. Office of Naval Research, the U.S. National Institute of Research, and DARPA (Defense Advanced Research Projects Agency).
Source: naftemporiki.gr
