Skip to content

Latest commit

 

History

History
56 lines (46 loc) · 1.67 KB

README.md

File metadata and controls

56 lines (46 loc) · 1.67 KB

MotionAplification

Motion magnification tools for offline videos and live camera streams.

Reference article: https://warped3.substack.com/p/motion-magnification

Features

  • Eulerian magnification for video files.
  • Live camera magnification with real-time preview.
  • Optional recording of processed live output.
  • Camera device dropdown with refresh.
  • Camera mode probing (resolution/FPS) and one-click apply.
  • Capture/save resolution controls.
  • Desktop UI (Tkinter), no web server required.

Requirements

  • Python 3.10+ recommended.
  • Windows/macOS/Linux with OpenCV-compatible camera/video codecs.

Installation

pip install -r requirements.txt

Run Desktop UI

python ui_app.py

Live Camera Workflow

  1. Open the Live Camera tab.
  2. Choose a camera from Camera Device.
  3. Click Probe Modes to detect common supported modes.
  4. Pick a mode and click Use Selected Mode (optional but recommended).
  5. Enable recording and choose Save As path if you want output saved.
  6. Click Start Live Camera.
  7. Press q in the OpenCV window to stop.

Video File Workflow

  1. Open the Video Processing tab.
  2. Select input video.
  3. Select output video path.
  4. Set magnification/filter parameters.
  5. Click Process Video.

Main Scripts

  • ui_app.py: desktop UI.
  • ampEul.py: file-based Eulerian magnification.
  • ampLive.py: live magnification pipeline.
  • ampPhase.py, ampPhaseSingleChannel.py: phase-based variants.

Notes

  • chunk_size must be greater than overlap.
  • high_cutoff must be greater than low_cutoff.
  • If a requested capture mode is unsupported by camera driver, OpenCV may silently fall back to a nearby/default mode.