Structured light measures 3D shape by projecting a known pattern and watching how it bends over surfaces — the high-detail depth tech behind the original Kinect, face-unlock, and precise short-range robot vision.
Structured light shines a known pattern (like a grid of dots) onto a scene. Where the pattern lands on something close, it looks different than on something far — and from how the pattern distorts, the sensor computes the 3D shape.
Instead of passively looking at a scene, why not paint it with a known pattern and watch how the pattern bends? That's structured light — one of the most accurate ways for a robot to see in 3D up close.
How it works
A projector casts a known pattern — stripes, or a dense grid of infrared dots — onto the scene. A camera offset from the projector sees that pattern distorted by the surface: it shifts and warps depending on how far each point is. From the geometry between projector and camera, triangulation converts that distortion into a depth value per point.
Project a pattern, read the distortion
Because the projected pattern is known, any deformation directly encodes surface shape — no need for the scene to have its own texture.
Why it beats stereo up close
Passive stereo vision fails on blank, textureless surfaces because it has no features to match. Structured light supplies its own texture (the pattern), so it reads even a plain white wall accurately, with fine detail. That makes it excellent for short-range, high-precision tasks — 3D scanning, bin-picking small parts, face recognition.
Structured light vs time-of-flight
Structured light — very accurate at close range and fine detail, but shorter range, slower, and easily washed out by sunlight (which drowns the projected pattern).
Time-of-flight — better at longer range and faster, but typically lower spatial detail up close.
The original Kinect used structured light (a projected IR dot pattern); later depth cameras and phone face-unlock systems use it or ToF depending on the range needed.
Where you'll see it
Depth cameras for indoor robots, industrial 3D inspection and scanning, robotic bin-picking, and face authentication. Its output is typically a point cloud for downstream perception and grasping.
Why it matters
Structured light gives robots dense, accurate close-range 3D even on featureless surfaces — a capability passive cameras can't match. It's a cornerstone of indoor depth sensing and precision inspection.
