LiDAR vs camera people counting

The right choice depends on the space. A camera counts from the image it sees, which suits defined entrances and service points. LiDAR measures distance with laser pulses, so it also works in the dark, outdoors and across wide-open areas, and it never takes a picture. A site with both kinds of space can use both.

The short version

• Cameras suit defined entrances, checkouts and service points, especially where cameras are already in place or planned.
• LiDAR suits dark, outdoor and wide-open areas, where one wide-area unit can cover what would take many overhead cameras.
• A counting camera analyses video on the device and sends on numbers; LiDAR never creates an image.
• Whether a setup passes a privacy review is for your own data protection officer to judge.
• A site with both kinds of space can combine the methods.

How does each method count people?

An AI people counting camera is a standard network camera with the counting built into the device. It analyses video as it arrives, recognises people crossing a line or occupying a zone, and sends on numbers: in/out counts, occupancy and queue metrics. Its close relative, the dedicated 3D people counting sensor, is a purpose-built overhead device using stereo vision or time-of-flight, mounted above a doorway for that single task.

LiDAR people counting never takes a picture. The sensor sweeps its surroundings with rapid, eye-safe laser pulses and times the reflections to build a three-dimensional map of everything that moves. People appear as anonymous clusters of points, and their position, height and direction of travel are enough to count them.

Which method suits which kind of space?

At a well-lit, defined entrance the two methods do the same job: exact in/out counts and occupancy. The differences show up when conditions get harder.

Does lighting matter?

A camera needs a scene it can read. Modern counting cameras handle a lot, but optical measurement is still tied to lighting conditions. LiDAR brings its own light. The laser works regardless of ambient lighting, at noon, at night or in total darkness, which helps in arenas after hours and in terminals before dawn.

What changes outdoors and in open spaces?

Forecourts, plazas, transport hubs and open-air sites combine changing daylight, weather and wide sightlines, which are hard conditions for an optical system. Lighting does not affect LiDAR there, but the hardware needs care: outdoor units must have housings rated for the weather and be placed where they survive rain, snow and direct sun. LiDAR measures moving geometry, so the same sensor family can also count vehicles and bicycles and measure parking occupancy. You get one measured picture from the door to the kerb. Radar copes with weather and darkness too, but it records vehicle presence and speed and gives a narrower picture where movement is mixed; traffic counting methods compares it with LiDAR.

How do they handle dense flows?

When people move close together, as in a surge at a gate or on a crowded concourse, a counting system has to tell individuals apart for the count to stay exact. LiDAR’s point cloud gives each person a distinct cluster of points, so people stay separate in the count even when they walk close together. A camera suits entrances where flows are channelled; LiDAR suits open, unchannelled spaces.

How many devices does the area need?

At a single doorway a camera is hard to beat, especially where cameras are already in place or planned, and an AI camera’s hardware can later extend into video security if you choose. Larger areas change that: in wide or open spaces, one wide-area LiDAR unit can cover what would take many overhead cameras, which keeps the device count down.

How do the two differ in a privacy review?

A counting camera analyses the video on the device and sends on only numbers. No personal images are stored and no individuals are identified. When a camera system counts people without identifying them, it can be part of GDPR-compliant footfall analytics, and that is how we run every camera deployment.

LiDAR makes no image of anyone, so a privacy review has no photograph or footage to assess. This matters where a visible lens can create friction: cultural institutions, public transit and other public space.

For a data protection officer, the practical difference is what you document: with a camera, that images are processed on the device and not stored; with LiDAR, that no image exists. Whether either setup passes is the judgement of your own data protection officer or compliance function. Every method we deliver is built with the same privacy-first engineering.

What if one site has both kinds of space?

Most sites mix several kinds of space. A transport hub has channelled gates and a wide-open concourse; a retail site has bright doorways and a dark car park. Choosing one method means accepting its blind spots wherever the space stops suiting it.

A hybrid setup combines the methods on one site: cameras at the defined entrances, checkouts and service points, and LiDAR across the dark, outdoor and wide-open areas.

Counts from each method reach you through one platform, one API and one set of dashboards, so every metric stays comparable whatever method produced it.

Bumbee Labs is the company behind the only footfall method in Europe approved by a data protection authority.

Sources

• Günter, A., Böker, S., König, M., & Hoffmann, M. (2020). Privacy-preserving people detection enabled by solid state LiDAR. 2020 16th International Conference on Intelligent Environments (IE), 1–4. https://doi.org/10.1109/IE49459.2020.9154970
• European Data Protection Board (2020). Guidelines 3/2019 on processing of personal data through video devices, version 2.0. https://www.edpb.europa.eu/documents/guideline/guidelines-32019-on-processing-of-personal-data-through-video-devices_en
• NOAA National Ocean Service. What is lidar? https://oceanservice.noaa.gov/facts/lidar.html