CrowdFaceYOLO: Advancing Real-Time Face Detection in High-Density Crowded Area
Waquar Ahmad, Aditya Vashist, Neel Sinha, Manisha Prasad, Vishesh Shrivastava, Junaid Hussain Muzamal · 2024
This research presents CrowdFaceYOLO, an advanced face detection model optimized for high-density crowds. Leveraging a dataset of 16.7 k images, the model incorporates custom modifications to the YOLO framework, achieving an impressive mAP of 0.8813 and an F1 Score peaking at ${0. 1 7 2 6}$. These results highlight the model’s superior precision and recall in crowded environments, establishing a new benchmark for face detection in challenging settings