WildARe-YOLO: A lightweight and efficient wild animal recognition model

Sibusiso Reuben Bakana, Yongfei Zhang, Bhekisipho Twala

Abstract: Highlights•Introducing a new, lightweight, and efficient animal recognition model.•Combines modified Stem Block, DWConv, and MBBlock for enhanced speed and accuracy.•Feature fusion enhanced with BiFPN for better accuracy.•CIoU loss was replaced with Focal-EIoU for class balance.•WildARe-YOLO – New approach in the limited infrastructure environment.