A Deep Learning Approach To Human Fall Detection With YOLOv12 Versi Nano
The original YOLOv12-nano model, known for its lightweight architecture, low latency, and efficiency on edge devices, was implemented on the Le2i dataset and achieved an mAP50 of 0.866 and an mAP50-95 of 0.555, providing a strong foundation for further improvement in elderly fall detection systems.