Resources
LabGym comprises three modules that create a cohesive workflow—enabling users to prepare, train, and analyze their behavioral data with accuracy and ease.
- Preprocessing Module: Optimize video footage for analysis—trim videos to focus on the necessary time windows, crop frames to remove irrelevant regions, enhance video contrast, reduce fps to increase processing speed, or draw colored markers to mark specific locations.
- Training Module: Customize LabGym to your research needs by training a Detector to locate animals or objects of interest and a Categorizer to recognize behaviors that you define.
- Analysis Module: After customization, perform automated behavioral analysis in videos/images. The module outputs comprehensive results and mines them for significant findings.
Getting Started
-
Installation Guide
step-by-step setup
-
Tutorial Series
video walkthroughs
Documentation
-
Practical Guide
action-oriented user guide
-
Extended User Guide
in-depth technical manual
Tools
-
LabGym Companion
AI troubleshooting & guidance
-
EZannot
AI-assisted annotation tool
U-M Great Lakes (HPC)
HPC system with Red Hat Linux provided to the University of Michigan.
-
Great Lakes HPC guide
instructions for installation + usage
Development
-
GitHub Repository
source code