LearnViz CompSci
Home / Learn / Graphics, Vision & Media / Optical Flow

Advanced Topic 42 of 49 Graphics, Vision & Media

Optical Flow

Estimating apparent motion between frames.

How it works

Optical flow estimates the apparent motion of pixels between consecutive video frames, producing a vector field showing where each point moves. Classical methods assume brightness constancy and smoothness: Lucas-Kanade solves for motion in small local windows, suitable for sparse tracking, while Horn-Schunck adds a global smoothness constraint for dense flow. Modern deep methods such as FlowNet and RAFT learn to predict dense flow and are far more accurate in difficult conditions.

Where it is used

Video stabilization, motion detection and tracking, action recognition, frame interpolation, video compression with motion estimation, robot navigation and driver assistance use optical flow.

Common pitfall

The aperture problem means motion along an edge cannot be determined from a local window. Large motions, occlusions, lighting changes and textureless areas break assumptions. Classical methods need image pyramids for large displacements. Flow is not the same as true 3D motion. Evaluate with endpoint error on benchmarks and validate with forward-backward consistency checks.

Where it fits in the course

Optical Flow is topic 42 of 49 in Graphics, Vision & Media, rated Advanced. It follows Image Classification. The next topic is Stereo Vision.

Practice and explore

How to study this topic

  1. Read the explanation above and restate the definition in your own words.
  2. Open the interactive lesson and answer its mini quiz.
  3. Read the common pitfall, then write a short example that avoids it.
  4. Check your understanding in the Graphics, Vision & Media quiz, and revisit the topic if you miss a question.

Frequently asked questions

What is Optical Flow?
Estimating apparent motion between frames. It belongs to Graphics, Vision & Media and is an advanced-level topic.
Where do people use Optical Flow?
Video stabilization, motion detection and tracking, action recognition, frame interpolation, video compression with motion estimation, robot navigation and driver assistance use optical flow.
What is the most common mistake with Optical Flow?
The aperture problem means motion along an edge cannot be determined from a local window.
What should I learn before this topic?
Start with Image Classification, the topic just before it in Graphics, Vision & Media.
How can I practice it?
Answer the mini quiz in the interactive lesson, then take the Graphics, Vision & Media quiz.
Is Optical Flow hard to learn?
It is an Advanced topic. Expect to read the explanation more than once and to connect it with earlier topics. Working through a small example by hand helps most.
How long does it take to learn Optical Flow?
Most learners can read the explanation and finish the interactive lesson in under an hour. Fluency comes from applying the idea, so plan a few short practice sessions with the quiz.
Which topics are related to Optical Flow?
Image Classification, Stereo Vision, Object Detection are closely connected. The full list is in the related topics section of this page.
What should I study after Optical Flow?
Continue with Stereo Vision, the next topic in Graphics, Vision & Media.

Related topics