▲ Jimin Roh, Integrated M.S./Ph.D. Student (left), and Professor Joongheon Kim (right), School of Electrical Engineering, Korea University
 

Jimin Roh, an integrated M.S./Ph.D. student in the AI and Mobility Lab (Advisor: Professor Joongheon Kim) at the School of Electrical Engineering, Korea University, has developed a low-complexity object detection technology for autonomous driving using Fast Quantum Convolutional Neural Networks (Fast QCNNs). By leveraging the advantages of quantum computing, this technology significantly improves the speed and efficiency of object detection in autonomous vehicles.
 

Object detection in autonomous driving has gained increasing attention with the advancement of deep learning, and continuous improvements in conventional convolutional neural networks (CNNs) have enhanced both performance and processing speed. However, as the complexity and volume of visual data continue to grow rapidly, traditional computing methods face limitations in computational speed.
 

To address this issue, the research team proposed a system called the Quantum Object Detection Engine (QODE), which implements a quantum version of CNN (QCNN) for object recognition. In particular, the team developed a new fast quantum convolution algorithm that enables the processing of multi-channel visual signals using a limited number of qubits.
 

QODE demonstrated that fast quantum convolution can drastically reduce computational complexity while maintaining effective object detection on multi-channel data. This advancement overcomes the current limitations of conventional QCNNs caused by the shortage of available qubits in real-world quantum hardware. Additionally, the team introduced a heterogeneous knowledge distillation training algorithm to further improve the performance of QODE.
 

This research was recognized for its significance and has been accepted for publication in IEEE Transactions on Mobile Computing, one of the most prestigious journals in the field of mobile systems, networking, and wireless communications. Jimin Roh's work focuses on developing novel autonomous driving algorithms based on quantum convolutional neural networks.