Structured Light Sensor Camera Explained: How SLS Cameras Detect a Figure

A structured light sensor camera, usually just called an SLS camera in ghost hunting circles, doesn't work anything like a normal camera. It doesn't capture light reflecting off a scene, it projects its own infrared pattern and reads how that pattern bends across every surface in the room. Here's what that actually means, and why a piece of motion-tracking hardware ended up as one of the most-used tools in paranormal investigation.

1.0What "Structured Light" Actually Means

Structured light is a known, fixed pattern of light, usually thousands of infrared dots, projected onto a scene from a fixed point. A sensor next to the projector reads that same pattern back. Because the sensor knows exactly what the pattern looked like when it left the projector, it can measure exactly how each dot shifted based on the distance and angle of whatever it hit. Do that across every dot in the pattern, and you get a full depth map of the room, not a photo, a measurement of distance at every point in the frame.

This is different from a stereo camera (which compares two lenses) and different from time-of-flight (which times a light pulse's round trip). Structured light gets its depth reading from pattern distortion instead, and it works in complete darkness since the "light" it's reading is infrared, not visible.

2.0From Depth Map to Skeleton Overlay

A raw depth map on its own is just a grid of distances, useful, but not something you can glance at and interpret. The second half of an SLS camera system is skeletal-tracking software that scans that depth data for the proportions and joint layout of a human body: head, shoulders, elbows, hips, knees, in roughly the arrangement a body holds them in. When it finds a match, it draws a stick-figure skeleton over that shape in the live feed.

That's the part people usually mean when they say "SLS camera", the combination of the depth sensor and the software drawing a skeleton over whatever it thinks is a humanoid figure, whether or not that figure is visible in a normal RGB image of the same room.

3.0Why the Hardware Wasn't Built for This

Every structured light sensor commonly used for ghost hunting, the Microsoft Kinect (V1 and V2) and the Intel RealSense D400 series, was designed for something else entirely: motion-controlled gaming, robotics navigation, and industrial 3D scanning. None of them shipped with a "detect a ghost" mode. What they shipped with was accurate, real-time human skeletal tracking in the dark, which turned out to be exactly the tool paranormal investigators wanted for a completely different reason: checking whether a room contains a moving figure nobody can see with their own eyes or a normal camera.

4.0What a Structured Light Sensor Can and Can't Tell You

• Can: detect a humanoid-shaped depth pattern in a scene, in complete darkness, in real time.
• Can: track that shape's position and rough posture as it moves.
• Can't: tell you what caused the shape. Pets, hanging objects, drapes catching air movement, and sensor noise can all occasionally produce a false skeleton lock.
• Can't: replace thermal or audio evidence. A skeleton with no correlated reading from anything else in the room is a weaker piece of evidence than a skeleton that lines up with an EMF spike or a cold spot at the same timestamp.

That last point is the real reason serious setups run an SLS camera alongside other instruments rather than by itself, a skeleton overlay is far more convincing when something else in the room reacted at the same moment.

5.0SLS Camera vs a Single-Sensor Setup

6.0Getting Started With SLS

If you already have a Kinect V1, Kinect V2, or Intel RealSense D435i sitting in a closet, that's a working structured light sensor, no purchase needed. SPECTER's free demo supports all three, running live neural-network entity tracking rather than the older single-skeleton overlay most SLS software ships with.

7.0Frequently Asked Questions