Quantum‐Soft Fusion

Real-time face detection is fundamental in various contemporary applications, including security, identity verification, biometrics, and human–computer interfaces. Therefore, even in the current state-of-the-art traditional and deep learning approaches, algorithms present difficulties in searching for a balance between accuracy, time for results, and computational complexity. Considering these issues, this study proposes the Quantum-Soft Fusion algorithm that leverages quantum computing for feature extraction and soft computing techniques for data management. Such integration helps the algorithm to surpass six competitive face detection approaches, with 98.2% accuracy, 96.5% precision, and a 96.1% balanced F1 score, and provides real-time results at 28 FPS. The requirement for this approach is attributed to the increasing need for precise real-time detection systems that can still achieve high performance while using limited computational power. The results shown here suggest that Quantum-Soft Fusion can improve the ACC and speed of face detection and propose a potential solution for next-generation face detection in real-life and real-scale applications.