Play Live Radio
Next Up:
0:00
0:00
Available On Air Stations
Watch Live

KPBS Midday Edition

Prize-Winning Roboticist Talks Self-Driving Cars In San Diego

Carnegie Mellon University professor Takeo Kanade is pictured on Nov. 12, 2016. He received the Kyoto Prize for Advanced Technology for his work on robotics and computer vision.
Inamori Foundation
Carnegie Mellon University professor Takeo Kanade is pictured on Nov. 12, 2016. He received the Kyoto Prize for Advanced Technology for his work on robotics and computer vision.
Prize-Winning Roboticist Talks Self-Driving Cars In San Diego
Prize-Winning Roboticist Talks Self-Driving Cars In San Diego GUEST: Takeo Kanade, 2016 Kyoto Prize laureate

A pioneer of automated driving is among the Kyoto Prize or its. This is a professor who has done research and development into computer vision technologies and robotics. He is in San Diego as part of the annual symposium which honors the prizewinners at universities across the city. It is my pleasure to welcome Takeo Kanadee . Welcome. Thank you. San Diego featured on May the car trips in 1985. Your team drove from Pittsburgh to San Diego using mostly hands-free technology. That must've been an extraordinary accomplishment at the time. You may call it extraordinary that is fun. It is not 98% with computer control. The rest, is the driver. They have to take over. Companies like Google and Tesla are a few years away from consumer regions of automated driver. How much does the current research go to your work? Back it is hard to say. I hope that at minimum, we showed it is possible. It is within the reach -- the reach of technology. You see, it took more than 20 years. Are you surprised about the length of time it has taken? Yes and no. When I was doing -- I felt it can be done tomorrow. So it took 30 years. I started the project in the mid-80s. More than 30 years. In that sense, it took too long. You have been working in the field of computer facial recognition for any years. How close are we speaking of taking a while, how close are we to facial recognition software being as reliable as a fingerprint? The facial -- the faith itself is not as recognized or reliable as a fingerprint. Even when we could do visually 100%, the question, cannot replace the fingerprint? The question may be -- if we replace it -- this is the question that is -- there is computer vision, can it be as good as a human? I would say yes. Absolutely. Is it today? Back I think it is almost. At least in some limited environment. What kind of limited environment? You should know that the human is not that good at recognizing that many people. Probably only 1000 people. That is it. If we do not recognize them -- if we do recognize them, we do not remember that at That is true. I have a hard time to remember people's face. Maybe that is the reason why our work on it. What fast -- fascinates you about robotics? Back it helps people -- It helps people. It makes our life easier and more comfortable and smoother. Especially, when it comes to people who may need some help to live independently like older people and people with some disabilities. Do you enjoy breaking down the way human beings go about doing things in order to be able to replicate them in computer vision or in an automated robotic sort of way? Back my philosophy is that it does not matter what computers should do in the way human does or can it do -- what should I do in different ways? My goal is to make it possible. I do not see any difference between the two. What are some of the real world problems you are trying to solve with robotics? Medical applications and surveillance problems and some bio engineering research. Research, they want to track the individual cells in the field of a microscope. That is tough. Even counting the number of cells us to when it comes to the task of tracking individual cells and when that particular cell divides itself and its daughters went where and divide further and so forth. It is impossible with the human but the computer can't. Your motto is think like an amateur, do like an expert. Why is it important to think like an amateur? I think the bottom line is, expert is the first expert and they are the first to say no, that is not possible. They know the difficulty. Amateurs tend to ignore them and simply approach the problem simply from what can be done. Estate for researchers and developers, they have to have that attitude. I am not saying researchers have to be amateur. No. Opposite. Researchers have to be expert that when you think of a problem, would you think of a solution, you have to get rid of ideas that it is difficult and how it should be done. That is why I say think like an amateur. I am speaking with Takeo Kanadee. He is a 2016 Kyoto Prize winner. He is giving a speech today. Think like an amateur, do as an expert. Professor, thank you. Think you. It is my honor to be on the program.

A pioneer in automated driving technology is among the Kyoto Prize laureates lecturing in San Diego this week.

Carnegie Mellon University computer science professor Takeo Kanade has done foundational research into computer vision technologies and robotics, including facial recognition software and computer-assisted surgical robots.

In November, he received the Kyoto Prize, an international award given by the Inamori Foundation "to honor those who have contributed significantly to the scientific, cultural, and spiritual betterment of mankind." Kanade's prize was in the advanced technology category.

Advertisement

Kanade is in San Diego as part of the annual Kyoto Symposium, which honors each year’s prize winners at universities across the city. He's speaking Wednesday at San Diego State University at 2p.m. about his motto: "Think like an amateur, do as an expert."

"When conceptualizing a problem and its possible solution, think simply and openly, as a novice in that field, without preconceived notions," Kanade wrote. "When implementing a solution, on the other hand, do so thoroughly, meticulously, and with expert skill."

Kanade joined KPBS Midday Edition on Wednesday to discuss the advances in robotics over his career and the self-driving car he helped develop that trekked from Pittsburgh to San Diego in 1995.