The rapid advancement of intelligent driving technology has attracted significant attention in both academic and industrial domains. Face detection technology is instrumental in advancing vehicle safety and optimizing the driving experience. This paper initially reviews the advancements in traditional face detection methods, then transitions to an analysis of several mainstream deep learning-based face detection techniques, including cascaded convolutional neural networks, two-stage, and single-stage detection algorithms. Furthermore, the study introduces lightweight face detection algorithms optimized for embedded systems, focusing on balancing real-time performance, accuracy, and computational efficiency in algorithms tailored for intelligent driving applications. Finally, potential avenues for future research are explored.