Computer Vision News - February 2019

detection and classification capabilities of the system. This hybrid solution combines Artificial Intelligence deep learning algorithms with advanced heuristic and non-neural networks to yield cutting edge classification in a very challenging environment, given by the variability of passengers’ looks and poses, the occlusions and the use of a wide-angle camera with a strong distortion effect. An additional challenge is given by the extreme variation in illumination and other spatial information, which is solved by pioneering sensor fusion technology. The Guardian sensor produces multiple channels of information which the system processes simultaneously by fusing those different data to support the complex decision. The obtained performance is almost real-time, since the system provides an all-inclusive report about the status of the car’s cabin in less than one second, regardless of all challenging conditions. “ In the last 6 months, I have had the pleasure of working with RSIP Vision on the development of computer vision algorithms ” says Yuval Gronau, Vice- President of R&D at Guardian Optical Technologies ; he adds that “ there were many technical challenges in the development of the algorithms and RSIP Vision came up with original solutions to overcome them. Thanks to this work, we have succeeded to significantly improve the performance of the algorithms. ” RSIP Vision’s CEO Ron Soferman concludes saying that “ the combination of Guardian’s advanced hardware with leading algorithms development is the ideal formula for the next generation’s systems. ” Project 13 Computer Vision News A project by RSIP Vision Take us along for your next Deep Learning project! Request a call here Feedback of the Month In the last 6 months, I have had the pleasure of working with RSIP Vision on the development of computer vision algorithms . There were many technical challenges in the development of the algorithms and RSIP Vision came up with original solutions to overcome them. Thanks to this work, we have succeeded to significantly improve the performance of the algorithms . Yuval Gronau Vice-President of R&D - Guardian Optical Technologies