Abstract
SUMMARYQuantifying movement is critical for understanding animal behavior. Advances in computer vision now enable markerless tracking from 2D video, but most animals move in 3D. Here, we introduce Anipose, an open-source toolkit for robust markerless 3D pose estimation. Anipose is built on the 2D tracking method Deep-LabCut, so users can expand their existing experimental setups to obtain accurate 3D tracking. It consists of four components: (1) a 3D calibration module, (2) filters to resolve 2D tracking errors, (3) a triangulation module that integrates temporal and spatial regularization, and (4) a pipeline to structure processing of large numbers of videos. We evaluate Anipose on a calibration board as well as mice, flies, and humans. By analyzing 3D leg kinematics tracked with Anipose, we identify a key role for joint rotation in motor control of fly walking. To help users get started with 3D tracking, we provide tutorials and documentation at http://anipose.org/.
Highlights
Tracking body kinematics is key to answering questions in many scientific disciplines
Methods for automated tracking of body kinematics from video have existed for many years, but they typically rely on the addition of markers to identify and disambiguate body parts
We introduce Anipose, a toolkit to quantify 3D body kinematics by integrating DeepLabCut tracking from multiple camera views
Summary
Tracking body kinematics is key to answering questions in many scientific disciplines. Neuroscientists quantify animal movement to relate it to brain dynamics (Mathis and Mathis, 2020; Seethapathi et al, 2019), biomechanists quantify the movement of specific body structures to understand their mechanical properties (Alexander, 2017; Bender et al, 2010), social scientists quantify the motion of multiple individuals to understand their interactions (Schwager et al, 2008; Halberstadt et al, 2016), and rehabilitation scientists quantify body movement to diagnose and treat disorders (Souza, 2016; Chiba et al, 2005; Rinehart et al, 2006) In all of these disciplines, achieving rapid and accurate quantification of animal pose is a major bottleneck to scientific progress. There is a pressing need for methods that perform automated, markerless tracking of body kinematics
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
