Community Effort

Animal Behavioral Dataset — Team Effort

Example clip from the behavioral dataset

We are building a large-scale, open animal behavioral dataset to support research at the intersection of neuroscience, behavior, and artificial intelligence. The dataset will combine rich videos of mice behaviors with annotations, enabling the development and benchmarking of new analysis tools such as TRACE.

This is a community effort, and we are actively looking for collaborators to join. We welcome contributors at every level — from undergraduate students to senior researchers — and from every scientific background.

What We're Building

  • High-quality multi-view video recordings of mouse behavior
  • Detailed behavioral annotations (social interactions, motor events, states)
  • Open benchmarks for behavior segmentation and classification models
  • Integration with neural recording data (Neuropixels, fiber photometry)

Who We Welcome

We welcome anyone interested in mouse behavior research, from all backgrounds, including:

Computer Science
Biology
Medicine
Neuroscience
Data Science / Statistics
Machine Learning / AI
Engineering
Related fields

Interested in Joining?

If you are interested in contributing to this effort, please email Dr. Zhang at guangwei.zhang@vcuhealth.org with a brief note about your background and what you'd like to contribute.

Email to Join the Team
Open-Source Tool

Social Behavior Annotator GUI

Social Behavior Annotator GUI
Screenshot of the Social Behavior Annotator (v1.2) — click to launch

We are developing an open, browser-based GUI for efficient annotation of social behaviors in mouse video recordings. The tool runs entirely in the browser — no installation required — and is designed to support the annotation workflow for our Animal Behavioral Dataset effort.

Launch the current development version directly from your browser and try it out. We welcome feedback, feature suggestions, and contributors interested in annotation tooling, UI/UX design, and computer vision.

Features

  • Runs entirely in the browser — no install needed
  • Frame-by-frame navigation with keyboard shortcuts
  • Custom behavior labels and multi-animal annotation
  • Export annotations in standard formats for downstream analysis

Contributing

Interested in contributing to the annotator development? Background in web development (JavaScript/HTML/CSS), computer vision, or behavioral neuroscience is especially welcome. Email Dr. Zhang at guangwei.zhang@vcuhealth.org to get involved.

Launch Annotator (v1.2) Contact About Development