A Chinese research team from the Shenzhen Institute of Advanced Technology (SIAT) has developed an AI technology called Social Behavior Atlas (SBeA) that addresses challenges in accurately detecting animal social behavior. SBeA can distinguish between similar-looking animals with over 90% accuracy, eliminating the need to define social behavior categories in advance. This technology could aid in discovering new, undefined differences in animal social behavior. The algorithm is effective in synthesizing large volumes of new data and training models with higher accuracy, providing more precise 3D social gesture estimation results. SBeA is suitable for various animals like mice, birds, and domestic dogs, with potential cross-species applications. The study was published in the journal Nature Machine Intelligence.
Source – CGTN