Turbulence Realm
Software Features How it works Tutorials Download Download
Get started

Download & install

Standalone executables for Windows and Linux — no Python installation or dependencies required. Just download and run.

Tracker · v3.0.0 · Production stable

Download Turbulence Realm-Tracker

A professional desktop application for tracking and analysing the movement of selected regions (e.g. bubbles) in fluid-dynamics videos. Installers bundle everything you need — no Python, no dependencies.

Choose your platform

Windows

Windows 10 / 11 · 64-bit · 78 MB

Download the setup installer, run it, and accept the no-liability disclaimer. Creates Start Menu shortcuts and an uninstaller entry.

Linux — .deb

Debian / Ubuntu · 64-bit · 141 MB

Recommended for Debian/Ubuntu. Installs to /opt, creates an Applications menu entry, and includes an uninstaller.

Linux — .run

Any distro · 64-bit · 186 MB

Self-extracting installer for any Linux distribution. Shows the disclaimer, installs to /opt, and creates a menu entry.

Install instructions

Windows (setup installer)

  1. Download TurbulenceRealmTracker-3.0.0-setup.exe.
  2. Double-click to run the installer.
  3. Read and accept the no-liability disclaimer on the license page.
  4. Choose install location and shortcuts.
  5. Launch from Start Menu → Turbulence Realm-Tracker.

Uninstall: Add/Remove Programs → Turbulence Realm-Tracker.

Linux (.deb — Debian/Ubuntu)

sudo dpkg -i turbulencerealmtracker_3.0.0_amd64.deb sudo apt-get install -f // fix any missing dependencies

Launch from Applications → Science → Turbulence Realm-Tracker or run TurbulenceRealmTracker in a terminal.

Uninstall: sudo dpkg -r turbulencerealmtracker

Linux (.run — any distro)

chmod +x TurbulenceRealmTracker-3.0.0-linux-installer.run sudo ./TurbulenceRealmTracker-3.0.0-linux-installer.run

Read and accept the no-liability disclaimer (type y). Launch from the Applications menu or run TurbulenceRealmTracker.

Uninstall: sudo /opt/TurbulenceRealmTracker/uninstall.sh

System requirements

  • Windows 10/11 64-bit or Ubuntu 20.04+/Fedora 36+ (x64)
  • Linux needs X11, OpenGL, and fontconfig
  • 4 GB RAM (8 GB recommended for large videos)
  • A display (the GUI requires a desktop environment)
  • 500 MB free disk space
Note: the installer bundles PyQt6, OpenCV, dlib, and all other dependencies. No Python installation is needed.

What you get

  • Video loading (MP4, AVI, MOV, MPEG, MKV) with frame navigation and trimming
  • Line, multi-point, and grid (checkerboard) calibration with presets and history
  • Bubble selection and tracking with real-time visual feedback and quality metrics
  • Live telemetry dashboard with velocity and displacement plots
  • Advanced analysis: smoothing, acceleration, FFT, cross-correlation, bootstrap CIs
  • Bubble classification and trajectory prediction
  • 3D trajectories, heatmaps, contact sheets, and time-lapse video export
  • Statistical reports (Markdown) and visualization export (PNG)
  • Session save/load with version history and annotations
  • Multi-video projects and batch processing
  • Auto-save recovery and lost-track re-acquisition
  • Dark/light theme, in-app documentation, and accessibility features
  • CSV, JSON, Excel, HDF5, and Parquet export
SINDy · v2.2.0 · MIT

Download Turbulence Realm SINDy

Video-based fluid-flow analysis with Sparse Identification of Nonlinear Dynamics. Extracts a dense velocity field, fits a SINDy model, predicts the reconstructed field, and provides rich visualisation, advanced analysis, and ML models. Installers bundle everything — no Python required.

Choose your platform

Windows

Windows 10 / 11 · 64-bit · 120 MB

Download the setup installer, run it, and accept the no-liability disclaimer. Creates Start Menu shortcuts and an uninstaller entry.

Linux — .deb

Debian / Ubuntu · 64-bit · 204 MB

Recommended for Debian/Ubuntu. Installs to /opt, creates an Applications menu entry, and includes an uninstaller.

Linux — .run

Any distro · 64-bit · 269 MB

Self-extracting installer for any Linux distribution. Shows the disclaimer, installs to /opt, and creates a menu entry.

Install instructions

Windows (setup installer)

  1. Download TurbulenceRealmSINDy-2.2.0-Setup.exe.
  2. Double-click to run the install wizard.
  3. Read and accept the no-liability disclaimer.
  4. Choose install location and shortcuts.
  5. Launch from Start Menu → Turbulence Realm SINDy.

Uninstall: Settings → Apps → Turbulence Realm SINDy → Uninstall.

Linux (.deb — Debian/Ubuntu)

sudo dpkg -i TurbulenceRealmSINDy-2.2.0-amd64.deb

The app appears in your application menu. Launch it or run /opt/TurbulenceRealmSINDy/TurbulenceRealmSINDy.

Uninstall: sudo dpkg -r turbulencerealm-sindy

Linux (.run — any distro)

chmod +x TurbulenceRealmSINDy-2.2.0-Linux-Installer.run ./TurbulenceRealmSINDy-2.2.0-Linux-Installer.run

The installer will display the no-liability disclaimer (type y to accept), ask for install location, create a menu entry, and offer a desktop shortcut.

Uninstall: /opt/TurbulenceRealmSINDy/install.sh --uninstall

System requirements

  • Windows 10/11 64-bit or Ubuntu 20.04+/Fedora 36+ (x64)
  • Linux needs X11, OpenGL, libGL, libEGL, and libglib2.0
  • 8 GB RAM (16 GB recommended for ML models and large videos)
  • A display (the GUI requires a desktop environment)
  • 1 GB free disk space
  • (Optional) NVIDIA GPU with CUDA for ML model training — CPU mode works for all other features
Note: the installer bundles PyQt6, OpenCV, PySINDy, and all other dependencies. No Python installation is needed. PyTorch (for ML models) is included with CPU support; GPU support requires a CUDA-enabled system.

What you get

  • Video loading with ROI selection and calibration
  • Optical flow extraction (Farneback, Lucas-Kanade, TV-L1, RAFT, PWC-Net)
  • SINDy equation discovery with polynomial, Fourier, combined, custom, and trig libraries
  • Optimizers: STLSQ, SR3, FROLS, constrained SR3
  • Cross-validation, model comparison, and time-delay embedding
  • Velocity field prediction and reconstruction
  • Analysis: vorticity, strain rate, divergence, spectra, POD, DMD, turbulence statistics, velocity PDFs
  • Physics-informed ML models (PINN, Autoencoder-SINDy, FNO, DeepONet, ConvLSTM, VAE, GAN, Ensemble)
  • Data quality: outlier detection, interpolation, noise estimation
  • Visualization: quiver plots, contours, streamlines, vorticity, animated heatmaps, POD/DMD modes
  • Export to CSV, HDF5, NetCDF, Parquet, JSON, PDF, and MP4 (FFmpeg included)
  • Full provenance metadata for reproducible results

Citation

If you use Turbulence Realm SINDy in your research, please cite it:

@software{rasheed_trsindy, author = {Rasheed, Fayaz}, title = {{Turbulence Realm -- SINDy}: Video-based Fluid Flow Analysis}, year = {2024}, license = {MIT} }

Disclaimer of Liability

This software is provided "AS IS" without warranty of any kind, express or implied, including but not limited to the warranties of merchantability, fitness for a particular purpose, and non-infringement. In no event shall the author be liable for any claim, damages, or other liability arising from, in connection with, or out of the use of this software.

By installing Turbulence Realm-Tracker or Turbulence Realm SINDy, you acknowledge and accept the no-liability terms. The full disclaimer is displayed during installation (Windows installer license page / Linux installer terminal prompt).

Downloaded? Learn the workflow.

Follow a tutorial and run your first analysis in minutes.

Browse tutorials →