Electrical engineers at the University of California, San Diego have developed a pedestrian detection system for smart vehicles and self-driving cars that spots pedestrians in real-time with an impressive accuracy.
Using deep-learning models and computer vision technology means the system can detect pedestrians at a rate of 2 to 4 frames per second – or roughly as well as the human eye can. To achieve reliable accuracy at this rate and in real-time, the algorithm filters out areas where human activity is not detected – for instance, the sky or the surrounding buildings.
Have you visited TNW's hype-free blockchain and cryptocurrency news site yet?
It's called Hard Fork.
Considering Nvidia’s decision to go into smart cars and Google’s interest in technologies for self-driving, smart and self-driving vehicles are likely to enter our lives in the near future, which is why detection algorithms will play a crucial role in maintaining order and safety on the streets.
Project lead, Prof. Nuno Vasconcelos, remarks that beside smart vehicles, the technology can also find application in robotics as well as image and video recognition systems.