Skip to content
Insights
Lidar is the 3rd generation of people counting and tracking technology

Meet the 3rd Generation of People Counting Technology

Mirroring the Technological Evolution Across Industries, People Tracking and Counting Technologies Have Now Entered Their Third Generation of Performance and Scope

Technological evolution is not a linear or uniform process. Significant improvements in performance and capabilities, which open up previously impossible possibilities, typically follow quantum leaps or technology generations.

The transition from approximate People Counting solutions to comprehensive Spatial Intelligence, based on continuous People Tracking, involves not just one, but two generational steps.

This evolution simultaneously also spans two axes, Performance and Scope, as summarized in the figure below and explained in more detail in this article.

LiDAR Technology is the key enabler of the 3rd technology Generation.

Lidar is 3rd generation technology for people counting and people tracking

LiDAR is the third generation of People Counting & Tracking technologies

Understanding the basics of 3D LiDAR Technology

Light Detection and Ranging, also known as LiDAR, is a technology for remote sensing that is used to measure distances in an environment.

Read article →

A generational change of Scope

A generational quantum leap occurs not only when a highly valuable capability, previously impossible with earlier technology, emerges, but also when this can be applied to a broader scope. This leap is also often accompanied by increases in performance.

1st and 2nd generation technology for people counting have privacy risks while the 3rd generation, Lidar, are natively anonymous

In the context of People Counting & Tracking, the key Scope aspects to consider are:

  • Can the technology accurately track actual individual people, as opposed to merely following proxies like their associated devices such as smartphones?
People counting is not just only tracking their smartphones

Counting smartphones is not counting people!

  • Global or Local application: is the People Counting & Tracking technology suitable for large-scale implementation across multiple areas (e.g. an entire airport across terminals or multiple levels in a retail store) or restricted to monitoring a specific spot (like passport control zones or just the footfall of a store) ?

In this example we show how Outsight’s technology based on LiDAR data can follow individual travellers across four different levels between a train station and an airport terminal over long distances, something not realistic with 2nd Generation technology:

Multi-level People Tracking

A unique feature of 3rd Generation tech is the ability, with the right software platform like Outsight’s, to individually follow the full journey of each individual visitor:

Each individual person can be tracked, among thousands of them

  • People Counting or Spatial Intelligence: does the technology primarily focus on just counting persons, or does it offer additional insights into their engagement with the surrounding environment, such as resource usage analytics?
People counting and people tracking KPIs that Outsight's 3D Spatial AI software measures

Privacy concerns are equally significant:

  • What potential risks exist regarding the protection of personal data?

Anonymizing data like Images is not the same as using a native Anonymous generation of technology, as explained here:

Anonymous vs. Anonymized : Learn the Difference

Understanding Anonymity in Sensor Data: discover the inherent privacy characteristics of each type of Sensor data and the potential risks associated with anonymizing sensitive information

Read article →

The third generation of People Counting and Tracking technology represents a genuine leap forward in terms of data privacy protection, offering a fundamentally different approach compared to previous generations.

Lidars are natively anonymous unlike cameras which have more privacy risks

Higher performance

As the technology evolves, each generation brings also significant performance improvements.

3rd generation people counting and people tracking technologies like lidar have the highest performance and purpose compared to the previous generations

As an example between Second Generation and Third Generation differences, a Camera cannot natively understand depth information nor the size, speed or volume of objects and people.

Camera's inability to understand depth

The lack of depth sensing directly relates to underperforming analytics

Stereo-vision cameras installed on ceilings encounter also several limitations in their ability to comprehend scenes due to their unique perspective and inherent optical constraints of the stereo-vision technique.

Getting close to 99% market expectations requires 3rd Generation-approaches that tackle the right problems to solve,

3D lidar detects object size & distance, volume, speed, and shape while cameras are only good at detecting color

The difference between Camera and LiDAR perception

To learn more how LiDAR technology compares to other solutions, take a look at the following article or download our People Counting Technologies Guide.

A detailed comparison of LiDAR, Radar and Camera Technology

This article explores the capabilities and limitations of each type of sensor, to provide a clear understanding of why LiDAR has emerged as a strong contender in computer vision tech race.

Read article →

Welcome to the 3rd Generation

Three conditions have been met for this new generation of technology to become widely accessible.

1. Accessibility: The necessary hardware is now performant and reliable.

There is now a diverse range of manufacturers offering various LiDAR sensor solutions, each with its unique strengths and weaknesses that cater to specific contexts.

This diversity allows for greater flexibility when selecting and combining the right models for a project.

For instance, Outsight’s customers typically integrate an average of three different manufacturers’ sensors per site, depending on their individual requirements.

2. Affordability: The cost is comparable to or lower than that of legacy technologies like Cameras.

The cost reduction stemming from LiDAR sensors is driven by a combination of factors. Firstly, the intense competition among numerous manufacturers fuels significant price drops in these advanced technologies.

Secondly, compared to second-generation solutions such as ceiling-mounted stereo-vision cameras, LiDAR sensors can cover between three and ten times more surface area per device.

This expanded coverage translates into substantial cost savings across various aspects of implementation. With fewer devices required per square meter, expenses associated with setup, wiring, networking, processing, and maintenance are significantly reduced.

These costs typically dwarf the expense of purchasing the sensors themselves, making LiDAR an economically attractive solution.

3. Proven & Scalable Usability: Spatial AI Software like Outsight’s not only has a proven track record but also offers the flexibility to handle large-scale projects.

Software specialists like Outsight, with extensive experience spanning many years, can also leverage the best hardware solutions from any manufacturer to deliver actionable insights.

Outsight's 3D Spatial AI software is compatible with all relevant lidar sensors

Beyond People Counting and Crowd Monitoring

The versatility of LiDAR extends beyond People Counting, powering solutions like Vehicle Tracking for smarter cities and roads, further highlighting its importance in advancing urban infrastructure.

Making Bellevue safer with LiDAR Solutions

Discover Outsight’s latest use case in the city of Bellevue, a complete LiDAR installation to increase users’ safety at busy intersections!

Read article →

Conclusion

Transitioning from approximate People Counting systems to advanced Spatial Intelligence solutions that rely on continuous People Tracking is a multi-faceted process, encompassing not one but two technological generations.

From transportation hubs such as airports and train stations, to quick service restaurants and retail stores, the most advanced companies are increasingly switching from 2nd to 3rd generation solutions based on LiDAR.

If you want to explore how we can help you with this transition, just let us know.

Related Articles

AIRPORTS

Aeroporti di Roma to deploy Outsight's Physical AI solution at scale across Rome Fiumicino Airport

Aeroporti di Roma (ADR) is expanding its collaboration with Outsight to a large-scale deployment across almost all Schengen common-use areas at Rome Fiumicino Airport.
CORPORATE

Intel and Outsight Announce Strategic Collaboration to Bring Physical AI–Powered Spatial Intelligence to the Enterprise Edge

Outsight’s Shift platform integrated into Google Distributed Cloud Edge powered by Intel Xeon 6 SoC – Live demonstration at Google Cloud Next 2026

Let's connect

Send us a Message

Drop your email and we'll get back to you as soon as possible.

Frequently Asked Questions

  • What is the difference between 2nd and 3rd generation people counting technology?

    Second-generation solutions, primarily ceiling-mounted stereo-vision cameras, measure depth indirectly through optical reconstruction and struggle with occlusion, lighting changes, and dense crowds. Third-generation systems use 3D LiDAR, which measures depth directly via laser pulses, classifies individual entities by size, shape, and speed, and tracks each person with a persistent anonymous ID across an entire site. Outsight's Motional Digital Twin represents this third generation in practice: infrastructure-mounted LiDAR sensors build a real-time, anonymous 3D replica of how every person moves through a space, deployed at scale across airports, train stations, and factories. The practical gap between generations is accuracy at scale: stereo-vision cameras cannot reliably resolve individuals in overlapping crowds, while LiDAR-based systems maintain centimeter-level spatial precision across thousands of concurrent entities.

  • Why does tracking smartphones give inaccurate people counts?

    Smartphone-based counting measures devices, not bodies. A single person carrying two phones registers as two entities, while a person with no phone is entirely invisible to the system. Device-based counts also suffer from consent and opt-in gaps, where devices are in airplane mode or have Bluetooth or Wi-Fi disabled. In transit environments with high device density and rapid turnover, the error rate can be large enough to render counts operationally useless for staffing or capacity decisions. This is precisely why infrastructure-based approaches using 3D LiDAR have gained traction: Outsight's Motional Digital Twin tracks physical shape and motion rather than signals, making every person visible regardless of what devices they carry or whether those devices are active.

  • How much more area can a LiDAR sensor cover compared to a stereo-vision camera?

    A single 3D LiDAR sensor mounted in infrastructure can cover between three and ten times the surface area of a ceiling-mounted stereo-vision camera. That ratio reduces device count per square meter, which in turn lowers costs for cabling, networking, edge processing hardware, and ongoing maintenance. Outsight builds on this principle across its deployments, using infrastructure-mounted LiDAR to cover large, complex sites, such as airport terminals and train stations, with fewer sensors than camera-based alternatives would require. Because secondary costs like installation and networking typically exceed the cost of the sensors themselves, the total-cost-of-ownership advantage of LiDAR over stereo-vision grows significantly beyond the hardware price comparison alone.

  • How many LiDAR sensor manufacturers does a typical site deployment use?

    Multi-vendor deployments are the norm rather than the exception. Sites integrating Outsight's SHIFT platform use sensors from an average of three different manufacturers per site. Different sensor models carry different strengths, such as range, resolution, field of view, or outdoor weather tolerance, so mixing models from competing manufacturers allows the configuration to match specific coverage requirements and cost targets at each zone within a site. Outsight's LiDAR-native architecture supports hardware from Hesai, RoboSense, Ouster, Velodyne, and Seyond, giving operators the flexibility to select the best sensor for each location without being locked into a single vendor's ecosystem.

  • Is anonymizing camera footage the same as using LiDAR for privacy compliance?

    These are structurally different approaches. Anonymizing camera footage means personal data, including faces or recognizable features, was recorded first and then processed to obscure it. Errors in that processing, or upstream data leaks before anonymization runs, create regulatory exposure. LiDAR captures geometry and motion only; no pixel image is ever produced, so there is no face or biometric attribute to anonymize in the first place. Outsight's infrastructure-based LiDAR platform operates on this principle by design: the Motional Digital Twin tracks the shape and movement of people and vehicles without ever generating an image, meaning personal data is never created rather than simply erased after the fact. Regulators and data protection officers treat this distinction as meaningful: one approach is a mitigation control applied after personal data exists, and the other is a system architecture where personal data is never generated.

  • Can 3rd generation people tracking follow individuals across multiple building levels?

    Yes, and cross-level continuity is one of the capabilities that distinguishes third-generation LiDAR-based tracking from earlier approaches. A software platform fusing sensors across floors can maintain a single persistent anonymous ID for a person as they move between a ground-level transit concourse and upper terminal levels, tracking the full journey without losing the entity at stairwells, escalators, or lift lobbies. Outsight's Motional Digital Twin applies exactly this principle in multi-level airport deployments, including Paris-Charles de Gaulle and Rome Fiumicino, where infrastructure-mounted LiDAR sensors hand off tracked entities continuously across levels. This end-to-end journey record enables dwell-time and route analytics that zone-by-zone counting systems cannot produce.