SLEAP: A deep learning system for multi-animal pose tracking

TLDR

Rapid advances in markerless pose estimation have enabled quantification of natural animal behavior, but extending these tools to multiple animals for studying social interactions remains challenging. The authors introduce SLEAP, a machine‑learning system designed for multi‑animal pose tracking. SLEAP offers a versatile workflow with a GUI, standardized data model, reproducible configuration, and over 30 model architectures, and it was evaluated on seven datasets spanning flies, bees, mice, and gerbils to compare part‑grouping, identity‑tracking, and architecture choices against existing methods. SLEAP achieves greater accuracy and speeds of more than 800 fps with latencies under 3.5 ms at full 1024 × 1024 resolution, enabling real‑time applications such as controlling one animal’s behavior based on social interactions with another.

Abstract

The desire to understand how the brain generates and patterns behavior has driven rapid methodological innovation in tools to quantify natural animal behavior. While advances in deep learning and computer vision have enabled markerless pose estimation in individual animals, extending these to multiple animals presents unique challenges for studies of social behaviors or animals in their natural environments. Here we present Social LEAP Estimates Animal Poses (SLEAP), a machine learning system for multi-animal pose tracking. This system enables versatile workflows for data labeling, model training and inference on previously unseen data. SLEAP features an accessible graphical user interface, a standardized data model, a reproducible configuration system, over 30 model architectures, two approaches to part grouping and two approaches to identity tracking. We applied SLEAP to seven datasets across flies, bees, mice and gerbils to systematically evaluate each approach and architecture, and we compare it with other existing approaches. SLEAP achieves greater accuracy and speeds of more than 800 frames per second, with latencies of less than 3.5 ms at full 1,024 × 1,024 image resolution. This makes SLEAP usable for real-time applications, which we demonstrate by controlling the behavior of one animal on the basis of the tracking and detection of social interactions with another animal.