Prof. Hiroshi Murase
Hiroshi Murase received the B.S., M.S., and Ph.D in electrical engineering from Nagoya University, Japan, in 1978, 1980, and 1987, respectively. In 1980 he joined the Nippon Telegraph and Telephone Corporation (NTT). From 1992 to 1993 he was a visiting research scientist at Columbia University, New York. From 2001 to 2003, he was Executive Manager of NTT Communication Science Laboratories. In 2003, he moved to Nagoya University. He is currently a Professor at the Graduate School of Informatics, Nagoya University. He was awarded the IEEE CVPR Best Paper Award in 1994, the IEEE ICRA Best Video Award in 1996, the IEEE Transaction on Multimedia Paper Award in 2004, etc. He got a Medal with Purple Ribbon from the Government of Japan in 2012. His research interests include computer vision, pattern recognition, multimedia information recognition, and ITS applications. He is a fellow of the IEEE, a fellow of the IPSJ, and a fellow of the IEICE.
Keynote Title: Image Recognition for Driving Assistance of Intelligent Vehicles
The driver assistance system through image recognition has become important, as the traffic accident rate of the elderly driver increases in Japan. This may also occur in other countries. On the other hand, the accuracy of image recognition has been improved by the development of deep neural networks or other machine learning technologies. However, a large size of learning data is necessary for these methods. First, I will present an efficient enhancement of learning data for in-vehicle camera images. Then, I will introduce several image recognition methods for driving assistance that we have developed so far, such as the recognition of traffic signs, the recognition of walking pedestrian while operating a smartphone, the recognition of the driver’s visibility, the weather recognition, and other research topics. In addition, I will talk about the human pose estimation using a super-low-resolution infrared sensor array for the purpose of watching the elderly people at home.