Online computational ethology based on modern IT infrastructure

In the study of animal behaviour, annotation and analysis is largely done manually either directly in field or from recordings. An emerging field, computational ethology, challenging this approach by using machine learning to automate process. However, use such methods general complicated a lack modularity, leading high cost long development times. At same time, benefits implementing fully automated pipeline are often minuscule. We propose online as way gain more automating process, making it easier ensure that equipment properly configured calibrated, enabling recording follow animals, even closed-loop experiments. work, we discuss requirements challenges for system an implementation based on modern IT infrastructure. Finally, demonstrate case studies bats mongoose. As algorithms developed expect systems enable new experimental setups insights field.