SLEAP (Social LEAP Estimates Animal Poses) is a multi-animal pose tracker based on deep learning. It is the successor of LEAP (Pereira et al., Nature Methods, 2019) and was designed to deal with the problem of tracking body landmarks of multiple freely interacting animals.
Using deep learning, SLEAP trains neural network models from few user annotations to enable highly accurate body part localization, grouping and tracking. It supports multiple neural network architectures, including pretrained state-of-the-art models and lightweight customizable architectures. SLEAP has been used successfully to track mice, fruit flies, bees and other species of animals under a variety of experimental and imaging conditions.
The software was designed to make it easy for users with no experience with deep learning through a fully featured GUI, as well as providing a rich functionality for advanced users seeking to develop a custom solution for their project. Tutorials and guides are available on our website (https://sleap.ai) detailing steps for easy installation (Windows/Mac/Linux), labeling a new project, training on the locally or on the cloud, and tracking new data.
Talmo Pereira; Joshua Shaevitz; Mala Murthy
This post was automatically generated by Talmo Pereira
- GuPPy, a Python toolbox for the analysis of fiber photometry data
- Computer-controlled dog treat dispenser
- Efficient training of mice on the 5-choice serial reaction time task in an automated rodent training system
- OPETH: Open Source Solution for Real-Time Peri-Event Time Histogram Based on Open Ephys