The project "Human Face Detection and Counting Using Advanced YOLOv26 and YOLOv12 Models" aims to develop a system for detecting and counting human faces in images or video frames. Using deep learning techniques, the project employs YOLO (You Only Look Once) models, specifically YOLOv26 and YOLOv12, to achieve accurate and real-time face detection. The system is designed to automatically identify human faces and provide a count of how many are present within a given image or video feed. The dataset used in this project is sourced from Roboflow, providing a diverse range of images with labeled human faces for model training. The solution utilizes the Flask framework for the backend, with a user-friendly front-end built using HTML, CSS, and JavaScript. This system can be implemented in various fields like security systems, crowd monitoring, and surveillance applications. By integrating advanced YOLO models, the project demonstrates the efficiency of deep learning in real-time face detection tasks. The approach ensures accurate predictions with high speed and minimal computational overhead, making it ideal for applications that require immediate responses. The proposed system will enhance the capabilities of face detection systems by providing both detection and counting functionalities.

Keywords: Face Detection, YOLOv26, YOLOv12, Deep Learning, Real-Time, Image Processing, Flask, Dataset, Human Counting, Surveillance.