SLS Camera Explained for Ghost Hunters: Verify the Figure First

An SLS camera is a depth-mapping sensor, usually a repurposed Kinect, that projects an infrared pattern or uses Time-of-Flight to build a 3D map of a room. Software then draws any human-like shape it finds as a stick figure. That figure is an algorithmic best guess, not a photograph, so you need to cross-check it before you call it evidence at HauntGears or anywhere else.

TL;DR:

• Sensors built from Kinect hardware work best indoors and within roughly 0.5 to 4.5 meters; sunlight, reflective surfaces, and competing infrared can create false figures.
• Depth frames and body masks show measured or segmented data, while skeletal figures are software estimates that can fail with occlusion or unusual poses.
• Change the camera angle, test a known person or object, and switch nearby infrared sources off and on to see whether the figure changes.
• Save raw depth or point cloud files when possible, and record the device, distance, and room layout for later review alongside other observations.

Table of Contents

• What does SLS mean, and what hardware sits inside these cameras?
• How does depth data turn into a stick figure?
• Why do SLS cameras show figures in empty space?
• How do you verify an SLS figure on-site?
• Why should you treat SLS output as one piece of evidence?
• Which HauntGears gear helps you back up SLS captures?
• FAQ
• Sources

What does SLS mean, and what hardware sits inside these cameras?

In ghost hunting, the term covers repurposed depth sensors, most often the Microsoft Kinect, that map 3D space instead of recording a normal image. Two methods do the mapping:

• Structured light: the sensor projects a pattern of infrared dots and measures how that pattern bends across surfaces. The bend tells it how far away each point is. Kinect v1 worked this way.
• Time-of-Flight (ToF): the sensor sends out infrared light and times how long it takes to bounce back. Kinect v2 moved to this method for better depth accuracy.

Kinect hardware was developed for gaming and can be used with software that looks for human bodies. That same design explains many failures in haunted buildings. The software can misidentify shapes when the scene differs from the conditions it handles reliably.

Expect these practical limits:

• Indoor use works best. Bright sunlight floods the sensor with competing infrared.
• Technical surveys put the reliable working range at roughly 0.5 to 4.5 meters, and noise grows outside it.
• Uneven textures and strong ambient light can produce noise that the software reads as movement.
• Newer Azure Kinect sensors use ToF too, but studies on validation artifacts show outside infrared sources can still interfere with measurements.

How does depth data turn into a stick figure?

The sensor never “sees” a person. It builds a stack of numbers, and software works through it in stages:

• Depth frame: a grid where each pixel holds a distance instead of a color.
• Point cloud: those distances converted into 3D points floating in space.
• Body mask: a cutout that separates shapes the software thinks are people from the background.
• Skeletal data: the final step, where joints and bones are fitted onto that cutout.

Depth frames and body masks represent sensor data, while skeletal data is a software interpretation. The skeleton is different. Motion-capture work shows that Kinect2 body-mask and depth streams can be more reliable than skeletal output when someone strikes a nonstandard pose. Skeletal models make assumptions, and they fail with occlusion or unusual body arrangements.

Skeletal tracking can make mistakes when input is noisy or incomplete. Motion-capture research notes that skeletal-model algorithms make assumptions and can fail with occlusion or unusual body arrangements, so a displayed figure is an interpretation rather than a direct image.

Depth-only analysis has strengths too. Research on fall detection shows depth frames can detect blobs and geometric features without recording faces, though the results depend on viewing angle. Multiple sensors may be used in some setups, but this article does not establish how common that is among handheld SLS units.

Pro Tip: If your unit lets you switch to the raw depth or body-mask view, review that footage first. It shows what the sensor measured before the software started guessing.

Why do SLS cameras show figures in empty space?

Most phantom figures trace back to a handful of causes. Work through them in this order:

1. Reflective surfaces. Mirrors, windows, glossy picture frames and polished floors change how the infrared pattern bounces back. Surveys list reflective surfaces and glossy materials as common sources of false positives.
2. Competing infrared. Sunlight through a window, a second SLS unit or an infrared night-vision illuminator can wash out or distort the pattern.
3. Range and field of view. Distances outside the sensor’s reported working range can add noise.
4. Occlusion. Occlusion can affect what the sensor captures.
5. Furniture and clothing. Furniture and other objects may contribute to ambiguous shapes in the captured data.
6. Awkward poses. Unusual poses can cause skeletal-model algorithms to fail or produce unreliable output.

When you review footage, consider whether environmental or sensor limitations could explain the figure.

How do you verify an SLS figure on-site?

Run a short routine while you are still in the room. Evidence is easiest to rule out, or keep, while the setup is untouched.

• Change the angle. Move the camera and repeat the capture. Compare captures from another angle to see whether the figure changes with the viewpoint.
• Add a known subject. Have a person or a familiar object step into the frame and compare how the software draws them.
• Toggle infrared sources. Turn nearby IR illuminators on and off, and shade windows. If the figure changes when the lighting changes, consider whether infrared interference may be affecting the capture.
• Cross-check with other tools. You can compare the capture with other observations, such as audio or readings from other equipment.
• Preserve the data. Save raw depth and point-cloud files when the device allows it. Note the device model, the distances involved and the room layout so later reviewers can judge the limits.

Placement matters as much as procedure, and our guide to static camera placement covers it in detail. For corroborating stills, see how to photograph ghosts. Once you have a visual match, a reverse image search workflow helps you confirm whether a capture is original or a known artifact.

Pro Tip: Log every check you ran, including the ones that failed to reproduce the figure. A clean record makes your capture more believable.

Why should you treat SLS output as one piece of evidence?

I treat an SLS camera as a mapping aid, not a witness. It tells me where something solid or reflective sits in a room, and it tells me nothing about what that thing is. The skeleton on screen came from a template, and templates fit noise all the time. So I describe captures conservatively, I lead with what I ruled out, and I never claim proof from one source. That is the same approach we take at HauntGears: safe, evidence-based investigation with plain explanations.

— Eric

Which HauntGears gear helps you back up SLS captures?

We built our picks around the verification checklist above. You can browse our SLS ghost-hunting camera and SLS camera category.

• GS2 Laser Grid System: projects a visible grid so you can check motion against a fixed reference.
• 4K Full Spectrum Camera Kit: high-resolution, full-spectrum capture to compare against a depth anomaly.
• ORDRO HDV-V12: infrared night vision for continuity checks across a session.

[image:cta_image]

Ready to build a full kit, from EMF meters to spirit boxes? Head to our shop and pick the tools that match your next investigation.

FAQ

How do you tell if a camera is recording at night?

Look for a recording indicator light or an on-screen timer, and check that the infrared illuminator glows faintly on a phone camera. If the unit has none of these, confirm by playing back a short test clip before you rely on it.

How do motion picture cameras work?

A lens focuses light onto a sensor, which captures a rapid series of still frames. Played back in sequence, those frames create the illusion of movement. An SLS camera works differently: it records depth data instead of ordinary light.

What is the working principle of a pinhole camera?

A pinhole camera lets light pass through a tiny opening onto a surface, projecting an inverted image without a lens. It has no depth sensing and no software, which is why it can’t produce SLS-style figures.

Can an SLS camera show a real ghost?

No sensor can confirm that. An SLS camera shows a software interpretation of depth data, so any figure should be checked against the environmental causes and cross-checks above.

Sources

• Kinect
• Kinect and RGBD Images: Challenges and
• A Practical Approach to Using Motion Capture in Performance Dance
• Minimizing Validation Artifacts in Azure Kinect

Recommended

• SLS Cameras for Ghost Hunting
• Static Camera Placement for Ghost Hunts: Setup Guide
• How to Photograph Ghosts: 4 Camera Setups & Settings
• Ghost Hunting Camera: Top Picks for 2026 Investigations

Share post