The recent advancements in the field of machine learning and computer vision have significantly enhanced the capabilities of real-time object detection, particularly in challenging environments such as low-visibility conditions and low-light scenarios. Innovations in model architectures, such as the integration of multimodal fusion techniques and the optimization of receptive fields, have led to substantial improvements in both accuracy and speed. Notably, the development of lightweight models like P-YOLOv8 and YOLO-TS has opened new possibilities for deployment on resource-constrained devices, making real-time detection more accessible and efficient. Additionally, the introduction of cloud-edge collaborative frameworks, exemplified by YOLO-Vehicle-Pro, addresses the computational demands of complex scenarios, ensuring robust performance in adverse weather conditions. These developments collectively push the boundaries of real-time object detection, contributing to advancements in autonomous driving, traffic safety, and cognitive workload management.