TRex, a fast multi-animal tracking system with markerless identification, and 2D estimation of posture and visual fields.

Tristan Walter,Iain D Couzin

doi:10.7554/elife.64000

Abstract

Automated visual tracking of animals is rapidly becoming an indispensable tool for the study of behavior. It offers a quantitative methodology by which organisms' sensing and decision-making can be studied in a wide range of ecological contexts. Despite this, existing solutions tend to be challenging to deploy in practice, especially when considering long and/or high-resolution video-streams. Here, we present TRex, a fast and easy-to-use solution for tracking a large number of individuals simultaneously using background-subtraction with real-time (60 Hz) tracking performance for up to approximately 256 individuals and estimates 2D visual-fields, outlines, and head/rear of bilateral animals, both in open and closed-loop contexts. Additionally, TRex offers highly accurate, deep-learning-based visual identification of up to approximately 100 unmarked individuals, where it is between 2.5 and 46.7 times faster, and requires 2-10 times less memory, than comparable software (with relative performance increasing for more organisms/longer videos) and provides interactive data-exploration within an intuitive, platform-independent graphical user-interface.

Highlights

Tracking multiple moving animals is important in various fields of research such as behavioral studies, ecophysiology, biomechanics, and neuroscience (Dell et al, 2014)
Results are saved to a custom, non-proprietary video format (PV) (Figure 2a)
Raw tracking speeds still achieved roughly 80 % accuracy per decision

Summary

Introduction

Tracking multiple moving animals (and multiple objects, generally) is important in various fields of research such as behavioral studies, ecophysiology, biomechanics, and neuroscience (Dell et al, 2014). Unbalanced assignments can be reduced to balanced assignments, for example using graph-duplication methods or by adding nodes (Ramshaw and Tarjan, 2012a, Ramshaw and Tarjan, 2012a). This makes the widely used Hungarian method (Kuhn, 1955; Munkres, 1957) a viable solution to both, with a computational complexity of Oðn3Þ. Re-balancing, by adding nodes or other structures, adds computational cost – especially when s ( n (Ramshaw and Tarjan, 2012b)

Methods

Results

Conclusion