Turbulence Realm turns ordinary fluid-dynamics videos into quantitative insight. Track bubble motion with sub-pixel precision and discover the governing equations of your flow — all on your desktop.
Purpose-built desktop tools that take you from raw video to publishable results — calibration, tracking, modeling, analysis, and export in a single workflow.
Track and analyse the movement of bubbles, droplets, and particles in fluid-dynamics videos. Real-time visual feedback, quantitative velocity and displacement data, and a full suite of analysis and export tools.
Extract a dense velocity field from video using optical flow, fit a SINDy model to discover the governing equations, reconstruct the field, and explore it with rich visualization and physics-informed machine learning.
A complete toolkit for fluid-dynamics video analysis — vision, modeling, analysis, and reproducible export.
Dense velocity fields and object tracking from video, with multiple backends and quality metrics.
Fit sparse models from polynomial, Fourier, and combined libraries. Read off the discovered governing equations directly, with cross-validation and library comparison.
Line, multi-point, and checkerboard grid calibration with presets and history.
Vorticity, spectra, POD/DMD, turbulence statistics, and statistical reports.
Train PINNs for Navier-Stokes, autoencoder-SINDy, Fourier Neural Operators, DeepONets, and ConvLSTMs — with uncertainty quantification and ONNX export.
CSV, JSON, Excel, HDF5, Parquet, NetCDF, PNG, PDF reports, MP4.
Provenance metadata — git commit, versions, seed, input hash — embedded in every export.
The same workflow powers both apps — load, calibrate, model, and export.
Open any MP4/AVI/MOV, draw a calibration line of known length, and get real-world units instantly.
Select regions to track, or let optical flow build a dense velocity field across the whole ROI.
Fit a SINDy model, run POD/DMD, compute turbulence statistics, or train a physics-informed net.
Quiver plots, heatmaps, animated fields, and reproducible exports to CSV, HDF5, PDF, and MP4.
Step-by-step guides for both apps — from your first tracking run to discovering governing equations.
Load a video, calibrate, select bubbles, track, and export your first velocity dataset — the full workflow.
Read →Improve accuracy with multiple reference lines and automatic checkerboard detection.
Read →Extract a velocity field, fit a SINDy model, and read off the discovered governing equation.
Read →Smoothing, FFT, bubble classification, trajectory prediction, and statistical reports.
Read →Train a PINN for Navier-Stokes and produce quiver, contour, and animated heatmap visuals.
Read →Installers for Windows and Linux — no Python installation required. Each installer includes a no-liability disclaimer, shortcuts, and an uninstaller.
Track and analyse bubble motion in fluid-dynamics videos. Setup installer for Windows; .deb and .run installers for Linux.
Video-based flow analysis with sparse identification of nonlinear dynamics. Setup installer for Windows; .deb and .run installers for Linux.
No Python or dependencies needed — the installers bundle everything. A display is required for the GUI.
The suite is ready for your footage. Start with a tutorial or grab the download.
Browse tutorials →