📰 01 paper is accepted at The 15th International Conference on ICT Convergence (ICTC 2024), Jeju, Korea.

Our paper titled "Towards Real-time Vietnamese Traffic Sign Recognition on Embedded Systems" has been accepted for presentation at the 15th International Conference on ICT Convergence (ICTC 2024) in Jeju, Korea. This work introduces the Vietnamese Traffic Sign Detection Database 100 (VTSDB100), an extensive dataset of 100 classes of traffic signs captured in diverse conditions in Ho Chi Minh City. We conduct a comprehensive evaluation of recent object detection architectures, including YOLOv8, YOLOv9, YOLOv10, YOLOX Nano, RetinaNet, and Faster R-CNN, to establish baseline performance on VTSDB100. The study also presents an optimized deployment pipeline on the Jetson Nano 2GB using TensorRT and quantization, achieving up to 32 FPS for real-time inference. Our findings highlight the feasibility of high-accuracy, real-time traffic sign recognition on resource-constrained embedded systems, providing valuable insights for future AI-powered driver assistance solutions.

• Phuong-Nam Tran, Nhat Truong Pham, Nam Van Hai Phan, Duc Tai Phan, Cuong Tuan Nguyen, and Dr. Duc Ngoc Minh Dang, “Towards Real-time Vietnamese Traffic Sign Recognition on Embedded Systems”, The 15th International Conference on ICT Convergence (ICTC 2024), Jeju, Korea