Animal Behavior

Animal behavior research aims to understand and quantify animal actions, interactions, and internal states, often leveraging advancements in computer vision and machine learning. Current research heavily utilizes deep learning architectures like convolutional neural networks (CNNs), transformers, and recurrent neural networks (RNNs), along with techniques such as instance segmentation and particle filtering, to analyze video and sensor data from diverse sources (drones, wearable tags, etc.). This work is crucial for improving animal welfare, advancing ecological understanding, and informing applications in areas such as precision agriculture and robotics, particularly through the development of automated behavior recognition systems.