A developer of robotics and artificial intelligence (AI) has been awarded a patent for a method that makes use of robots in order to train visual and three-dimensional properties onto computer vision models.
At the most recent AI Week event, which took place in Tel Aviv, the Israeli business Deep Learning Robots, which has its headquarters in the city of Binyamina-Giv’at Ada, revealed its brand-new patent. Since 2016, when it first created and connected visual and spatial information automatically and in real-time, the firm has been a pioneer in the use of robots to provide correct training data. In 2016, the company also linked visual and spatial elements automatically.
According to the firm, DLR’s technologies helped lead to more accurate models and enabled faster deployments of robotic vision systems. Furthermore, the company believes that these systems have the potential to someday transform the way that systems create 3D data. CEO Carlos Benaim stated that the technology may also make it possible for robots to effectively discover and identify things even when they are just equipped with simple, two-dimensional camera systems.
The firm has stated that the patent provides a “testament” to its dedication to pushing the frontiers of robotics and that the company is “eager to continue to explore the full potential of this technology.”
DLR asserts that it is a pioneer in the development of innovative solutions for automation, including a robotic controller that enables a robot to learn tasks just by witnessing people carrying them out.