TropiCam-AI Tracks Tree-Dwelling Species with 95% Accuracy

Scientists developed TropiCam-AI, the first deep learning algorithm for identifying arboreal vertebrates in the Neotropics ^[1]. This new model classifies 84 taxa, including 63 specific species of mammals and birds ^[1][2]. Unlike previous tools, this technology focuses on animals living in the forest canopy ^[1].

The system achieves a high 95% accuracy rate for species classification ^[1]. Researchers from the University of Rome La Sapienza designed the tool to address a critical gap in monitoring ^[2]. Most existing wildlife AI systems focus on ground-dwelling animals rather than those in the trees ^[1].

This innovation allows conservationists to better track biodiversity in complex tropical environments ^[1]. By automating the data processing of arboreal camera traps, the project speeds up essential research ^[2].