Measuring distance directly instead of inferring it
Stereo vision infers depth indirectly, by comparing two images and doing geometry. An alternative approach is to measure distance directly with dedicated hardware. LiDAR, short for Light Detection and Ranging, fires rapid laser pulses out into the scene and measures the time it takes for each pulse to bounce off a surface and return to the sensor. Since light travels at a known, constant speed, that round-trip time converts directly into a distance, a technique called time-of-flight. Sweeping this measurement across many directions builds up a set of 3D distance readings covering the scene.
Structured light sensors take a different direct-measurement approach: they project a known pattern, often a grid or set of dots, onto the scene using infrared light, then observe how that pattern deforms when it lands on surfaces at different distances and angles. A flat surface straight ahead reflects the pattern with minimal distortion, while a surface that is angled or far away warps the pattern in predictable ways that a matched sensor can decode back into depth. Both LiDAR and structured light skip the correspondence-matching problem that stereo vision faces entirely, since they are not comparing two camera images, they are measuring a physical signal.