Review: Tiny Face Detection and Recognition Techniques

The exponential growth of video and image databases has created a need for intelligent systems to automatically analyze information, as manual efforts are no longer feasible. Faces plays a crucial role in social interaction, conveying identity and emotions, thus requiring efficient and accurate analysis. Deep learning techniques has brought about a significant revolution in face detection, despite their increased computational requirements. This paper presents a comprehensive analysis of representative deep learning-based methods for face detection, focusing on their accuracy and efficiency. It also compares and discusses popular and challenging datasets, including their evaluation metrics. Additionally, a thorough comparison of successful deep learning-based face detectors is conducted, evaluating their efficiency using Floating Point Operations (FLOPs) and latency as metrics. The results and findings of this study can serve as a valuable guide for selecting suitable face detectors for various applications. Moreover, they can contribute to the development of more efficient and accurate detectors. The paper aims to address the pressing needs for intelligent systems that can automatically understand and analyze visual information in an increasingly data-driven world.