# Point Cloud

> A 3D scene stored as many dots. Each dot is one spot in space.

**Field:** Artificial Intelligence  
**Category:** Robotics & Vision  
**Also known as:** 3D Point Cloud, Point Cloud Data

## Think of it like

It is like a picture drawn only with dots. Close up you see dots, but step back and you see a room.

## Definition

A point cloud is a 3D picture made of many dots. Each dot is one spot in space, stored as three numbers for its place. Some dots also carry a colour or a strength. Lidar and depth cameras make point clouds, often with many thousands of dots in one scan. A robot uses them to find walls, floors and objects. Software can also join the dots into a solid 3D model.

## Example

Lucy scans a room with a lidar on a tripod. She gets a point cloud of the walls and furniture, and uses it to plan new shelves.

## Why it matters

Point clouds are how many robots and self-driving cars see shape and distance. Mapping and finding objects in 3D starts here.

## Common mistake

Treating a point cloud like a photo. It has gaps where the sensor could not see, and it has no surfaces until software adds them.

## Three principles

- **Dots in space.** Each dot has three numbers.
- **From scanners.** Lidar and depth cameras make them.
- **Gaps and noise.** Hidden spots have no dots.

## Related terms

- [Light Detection and Ranging (LiDAR)](https://learn.benibix.com/term/ai/lidar)
- [Depth Estimation](https://learn.benibix.com/term/ai/depth-estimation)
- [Simultaneous Localization and Mapping (SLAM)](https://learn.benibix.com/term/ai/slam)
- [Sensor Fusion](https://learn.benibix.com/term/ai/sensor-fusion)
- [3D Modeling](https://learn.benibix.com/term/design/3d-modeling)

![Three panels. Left: a lidar sends a flash of light to a wall and the echo comes back. Middle: a wall, a floor and a chair drawn only as dots, a point cloud. Right: a camera photo beside its depth map, where the near box and the far wall get different shades.](https://learn.benibix.com/diagrams/ai6-3d-sensing.svg)

---

Source: https://learn.benibix.com/term/ai/point-cloud
Author: Behnia Zarei (a developer in Vienna, Austria, who builds and teaches data, AI and business intelligence at BENIBIX)
Publisher: BENIBIX Learn (https://learn.benibix.com/)
Last updated: 2026-11-17
Licence: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Quote freely with attribution.
