Back to all work
Research · i2bl Lab, UCLA

Ferrofluid Computer Vision

Real-time vision for closed-loop magnetic control · Nov 2025 – Present

I'm developing a real-time computer-vision system that segments and localizes ferrofluid flow and magnetic sources from a live camera feed, then closes the loop, feeding those detections back to algorithmically place the magnetic field and steer ferrofluid trajectories across a microfluidic chip.

Computer VisionReal-time InferenceClosed-loop ControlMicrofluidics
The loop

Vision in, magnetic field out

The system runs as a perception-to-actuation loop: see the fluid and the sources, decide where the field should go, move it, and watch the result, continuously.

Sense
Live Camera
microfluidic chip
Segment
CV Model
ferrofluid + sources
Localize
Spatial Detect
positions in frame
Actuate
Field Placement
algorithmic
Result
Trajectory
controlled flow
Contributions

What I'm building

  • Trained a real-time CV model to segment and localize ferrofluid flow and magnetic sources from live video.
  • Integrated the spatial detections into an algorithm that adjusts magnetic field placement for precise correction and control.
  • Targeting reliable control of ferrofluid trajectories on microfluidic chips.
Stack

Tools

PythonComputer VisionReal-time SegmentationMachine LearningClosed-loop ControlMicrofluidics

Have a demo clip or figure from the lab? It can drop right in here.