Professor Jae Wook Jeon's team Wins First Place in AI City Challenge at CVPR

-Envisioning Future Intelligent Traffic Surveillance System Utilization

In June 2023, Professor Jae Wook Jeon’s team from the Department of Semiconductor Systems Engineering achieved 1st place in the 7th AI City Challenge, titled "Detecting Violation of Helmet Rule for Motorcyclists," hosted by CVPR (Computer Vision and Pattern Recognition), the world's leading conference in the field of artificial intelligence and computer vision, which took place in Vancouver, Canada.

In this inaugural competition held in 2023, a total of 39 teams from around the world participated, competing from January to June over a span of five months to develop technology to find out whether motorcycle riders were wearing their helmets.

Professor Jeon Jae Wook's team devised an artificial intelligence system with a two-stage structure. First, they identified the positions of various motorcycles and their riders. Then, they detected whether the riders were wearing helmets. This innovative approach earned them the highest detection performance score.

This victory has allowed Professor Jae Wook Jeon's team to showcase their technological prowess on the global stage. The developed technology is expected to find widespread application in future intelligent traffic surveillance systems.

The technology, led by graduate students from the Department of Electrical and Computer Engineering participating in the BK21 program, is the result of the research project "Real-time Intelligent Traffic Surveillance System on Edge Devices." This project was conducted under the auspices of SW StarLab (Software StarLab), an initiative by the Ministry of Science, ICT, and Future Planning and the National IT Industry Promotion Agency to secure global software technology and cultivate master's and doctoral-level talents in the field of software.