Skip to main content

CV Zebrafish

Overview​

CV Zebrafish is a desktop application that automates the analysis of zebrafish movement data from DeepLabCut tracking outputs. It replaces a manual process that takes 2–3 hours per dataset with an automated pipeline that completes in roughly 5 minutes. The tool validates DeepLabCut CSV files, auto-generates analysis configurations, computes kinematic metrics (fin angles, head yaw, tail dynamics, swim bouts, spine angles), and renders interactive Plotly graphs — all through a guided PyQt desktop interface. It also supports cross-correlation analysis between body part movements and multi-dataset comparison.

Information​

  • Source Code: https://github.com/oss-slu/cv_zebrafish git
  • Client: Mohini Sengupta, Ph.D.
  • Track: Client-driven Product
  • Current Tech Lead: Madhuritha Alle github
  • Developers:
    • Bruce Miller (capstone) github
    • Kwabena Adjei Omanhene-Gyimah (capstone) github
    • Nilesh Gupta (alumni) github
    • Gihwan Jung (alumni) github
    • Jacob Winter (alumni) github
  • Start Date: Aug 11, 2025
  • Technologies Used:
    • Python 3.10, PyQt5 (desktop GUI)
    • NumPy, Pandas, SciPy (data processing and kinematic calculations)
    • Plotly, Kaleido (interactive graphs and static image export)
    • OpenCV (computer vision utilities)
    • SQLite (session and run persistence)
    • Conda (environment management)
  • Type: Desktop Application
  • License: MIT

Technical Information​

Development Priorities​

  • Integrate multi-CSV comparison results and cross-correlation display into the Graph Viewer UI
  • Add interactive graph controls (clickable points, local extrema capture)
  • Validate angle calculations across new dataset formats
  • Set up CI/CD pipeline with GitHub Actions
  • Modernize application interface and interactivity

Get Involved​

If you would like to contribute to this project, please visit our GitHub page to create issues or pull requests.