Data and Tools

From data to discovery

Behavior has become a big data science. At MPI-AB, our mission is driven by data-oriented research with an emphasis on creating and making available the next generation of quantitative tools. Our researchers are supported by advanced computing technologies and innovative software solutions that enable them to turn their high-volume data streams towards answering the complex questions of modern behavioral research. In turn, the data and code they produce is shared openly, helping the global research community make the discoveries of tomorrow.

Movebank is a free, online database of animal tracking data hosted by MPI-AB. Movebank helps animal tracking researchers to manage, share, protect, analyze, and archive their data. more
The Animal Tracker App draws animal movement data out of Movebank and shares it with the global public on a user-friendly interface. Users can follow tracked animals wherever these happen to be in the world, while in turn contributing to research by posting observations and photos of tagged animals that they encounter. more
In the MoveApps project we are creating a platform to make sophisticated analytical tools accessible to a larger audience. Users are invited to combine multiple analysis Apps to create a Workflow that analyses data from the Movebank database. The Apps are created by the MoveApps team or the broader coding community in R, R-shiny or other languages. Our goal is to bring together movement ecologists and wildlife managers looking for flexible analysis tools with those who are developing tools and want to share them with others. more
TREX can track up to 100 individuals while maintaining their separate identities without the need for physical markers. The tool also integrates features such as estimating an individual’s 2D posture and visual field, and the ability to track positions of up to 256 individuals in real-time. more
DeepPoseKit is a software toolkit with a high-level API for 2D pose estimation of user-defined keypoints using deep learning—written in Python and built using Tensorflow and Keras. more
Open source tools from computer vision and pose estimation to modelling collective behavior. more
For decades, researchers have been observing orangutans at Suaq Balimbing in Indonesia and compiling one of the largest databases on orangutan behavior. This database will pave the way for understanding how skills and abilities develop in individuals and, ultimately, the evolution of high-level cognition in great apes. more

Go to Editor View