Most wildlife AI focuses on the ground. This model looks up in the trees
Most wildlife AI focuses on the ground. This model looks up in the trees
Publish Date: 2026-05-28 02:47:00
Source Domain: news.mongabay.com
- New AI Model for Arboreal Species Detection: Scientists have developed TropiCam-AI, an artificial intelligence model specifically designed to detect and identify tree-dwelling species in tropical forests of the Americas.
- Focus on Arboreal Species: Unlike other AI models that concentrate on ground-dwelling animals, TropiCam-AI aims to fill the gap in the detection of arboreal species, which are crucial for ecosystem functions as seed dispersers and yet often overlooked.
- High Detection Accuracy: The TropiCam-AI can recognize 84 taxa, including 63 different species, with an accuracy rate of 95%. Most taxa achieve over 90% precision and recall, demonstrating the model’s effectiveness.
- Application in Camera-Trap Data Analysis: The model can analyze camera-trap data, helping ecologists and researchers expedite the processing of millions of images and videos, aiding in conservation efforts for tree-dwelling species threatened by deforestation.
- Data Collection and Training: The bulk of data for training the model was gathered by collecting camera-trap images from expeditions in Brazil and supplementing with data from other locations and citizen scientist platforms like iNaturalist.
- Collaborative Development: The project involves collaborative efforts between researchers, local communities, NGOs, and multiple international sites to enhance the model’s training data and expand its capabilities.
- Future Prospects: The development team plans to refine the AI model further by adding more training data and working with additional collaborators to enhance its precision and applicability.
