1. Why Do Retailers Need More Than Basic People Counting?
Traditional people counting systems address a fundamental question: how many people pass through a store entrance?
This information helps retailers understand visitor volume, identify peak periods, and compare traffic patterns. However, for businesses that want to understand actual customer visits, a basic entry count may not tell the whole story.
Consider a store that records 1,000 entry events in one day. Does that mean 1,000 different customers visited? Not necessarily. Employees may enter and leave several times, delivery personnel may pass through the entrance, and the same shopper may visit more than once.
Without appropriate identification and counting rules, these events can make it difficult to interpret the actual customer traffic.
At FOORIR, we believe people counting should go beyond recording numbers. It should help retailers understand the movement patterns and visitor activity behind those numbers.
That is the thinking behind our HX-CCD21. By combining 3D binocular vision, AI-based human detection, and multi-dimensional traffic analytics, we aim to provide retailers with a more detailed view of store traffic.
2. Five Key Differences Between HX-CCD21 and Traditional People Counters
People counting systems use different detection technologies. Infrared sensors, beam counters, conventional vision systems, and 3D vision devices each have their own applications.
Rather than treating all traditional systems as the same, we focus on the capabilities that matter to retailers: how people are detected, how repeated observations are handled, and what additional information can be extracted from traffic data.
2.1 Dual-Lens 3D Vision: Adding Spatial Information to People Detection
Some basic people counters use infrared or photoelectric sensors to detect changes when someone passes through a defined entrance area. These solutions can be effective when traffic boundaries are clear and counting requirements are relatively simple.
Our HX-CCD21 uses dual-lens 3D binocular vision. Two lenses capture images from different viewpoints, and the system uses image disparity to estimate depth.
Compared with approaches that rely only on two-dimensional image information or simple detection signals, depth information provides an additional basis for understanding human positions and movement.
At retail entrances, this helps support human detection and directional people counting.
Of course, 3D vision does not eliminate every counting challenge. Entrance layout, occlusion, crowd density, and installation conditions can still influence real-world performance.
2.2 AI Traffic Analytics: Looking Beyond Entry and Exit Counts
Basic people counting generally focuses on the number of people passing through, entering, or leaving a monitored area.
We want to provide retailers with more ways to examine store traffic.
Our HX-CCD21 supports several AI-related functions:
- Entry, exit, passing, staying, and returning measurements.
- Age and gender analysis.
- Employee identification.
- ReID-based visitor deduplication.
- Dwell-time analysis.
These measurements can help retailers examine traffic changes, visitor movement, and activity across different periods.
Age and gender outputs are algorithmic estimates intended for aggregate traffic analysis. They should not be interpreted as definitive personal attributes for every individual visitor.
2.3 ReID Deduplication: Reducing Repeated Visitor Records
In real retail environments, the same shopper may pass through an entrance several times or appear in different monitored areas.
If every detection is treated as a separate visitor, the resulting data may contain repeated records.
To address this challenge, we integrate AI-based Re-Identification, or ReID, into the HX-CCD21.
ReID uses human feature information to associate detections across different times or monitored areas. This helps reduce repeated records associated with the same visitor.
For example, in a multi-entrance commercial environment with the appropriate configuration, the system can attempt to associate detections from different monitored areas to support visitor deduplication.
ReID is an algorithmic association process, not real-world identity verification. It cannot guarantee that every observation will be matched correctly under all conditions. Occlusion, lighting, and deployment configuration can affect performance and matching scope.
2.4 Employee Identification: Reducing the Impact of Staff Movement
Retail employees may enter and leave a store many times throughout a working day. If all these movements are included in customer traffic totals, the resulting data may not accurately represent customer visits.
Our HX-CCD21 supports employee identification through employee badge recognition, helping distinguish staff activity from other visitor movement and filter employee entry and exit records.
This allows retailers to reduce the impact of internal staff movement on customer traffic measurements.
Employee identification and ReID deduplication are two different functions:
- Employee identification helps distinguish employees from other people.
- ReID deduplication helps associate repeated human detections.
They complement each other, but they have different purposes and configuration requirements.
2.5 From Counting People to Understanding Store Operations
We designed the HX-CCD21 to do more than generate another counting report. Our goal is to provide traffic information that retailers can use to investigate real operational questions.
For example, store managers can examine:
- Entry traffic across different time periods.
- Daily and weekly traffic peaks.
- The influence of employee movement on visitor counts.
- Changes in visitor dwell time.
- Traffic differences between store locations.
- Relationships between traffic patterns and sales data.
It is important to remember that higher footfall does not automatically mean higher sales. We encourage retailers to combine traffic measurements with POS transactions, operating hours, and campaign data rather than relying on visitor numbers alone to evaluate store performance.
3. How Do the Technologies in HX-CCD21 Work Together?
The value of the HX-CCD21 does not come from one isolated feature. It comes from combining human detection, visitor association, and traffic analytics.
3.1 3D Vision Provides Spatial Perception
Dual-lens 3D vision provides spatial information for human detection. It helps the device understand people’s positions and movements, creating a foundation for further processing.
3.2 AI and ReID Support Further Analysis
After detecting people, AI algorithms process the relevant information.
ReID helps associate potentially repeated human detections, while employee identification helps distinguish internal staff movement from other visitor activity.
Together, these functions help reduce unnecessary interference in customer traffic measurements.
3.3 Traffic Analytics Turns Detection Results Into Operational Data
After detection and analysis, the system can provide different types of information, including entry and exit counts, dwell-time statistics, and age and gender analysis.
These measurements give retailers more ways to examine traffic patterns and support operational decisions.
4. Basic People Counting Solutions vs. FOORIR HX-CCD21
| Comparison | Basic People Counting Solutions | FOORIR HX-CCD21 |
|---|---|---|
| Primary purpose | Basic passage or entry counting | AI people counting and traffic analytics |
| Detection method | Varies: infrared, beam sensors, or vision | Dual-lens 3D vision and AI |
| Directional counting | Depends on the device | Supported |
| Age and gender analysis | Generally not included in basic models | Supported |
| Employee identification | Depends on device and configuration | Supported |
| Repeat visitor deduplication | Depends on system capability | ReID supported |
| Dwell-time analysis | Depends on system capability | Supported |
| Data applications | Mainly basic counting reports | Additional traffic analytics dimensions |
The term “basic people counting solutions” refers to common entry-level counting approaches. It does not mean that all other brands or technologies lack advanced capabilities. Actual features vary by device and configuration.
5. How Can We Help Retailers Gain More Value From Traffic Data?
The HX-CCD21 can be applied in a range of commercial and public environments. In retail, we see particular value in the following applications.
Chain Retail Stores
By collecting traffic data across multiple locations, retail managers can compare visitor patterns between stores and time periods. These insights can support operational planning and workforce scheduling.
Fashion and Apparel Stores
Age, gender, and dwell-time analysis can provide additional perspectives on visitor composition and time spent in the store. Retailers can combine these measurements with product display and sales information to evaluate store performance.
Convenience Stores and Coffee Shops
Understanding entry volumes and peak traffic periods can help operators review staffing arrangements and service resources throughout the day.
Shopping Malls and Multi-Entrance Commercial Spaces
With suitable device configurations and monitoring coverage, ReID deduplication can help reduce repeated visitor records across different monitored areas, providing additional context for traffic analysis.
6. What Should You Consider Before Installing HX-CCD21?
To achieve the expected results, we recommend evaluating the site conditions before installation.
Installation height: The HX-CCD21 supports an installation height range of 1.9–6.0 meters. For AI functions, we recommend an installation height of 2.5–4.0 meters.
Coverage: The supported coverage range is 1.1–8.4 meters. Actual coverage should be evaluated according to installation height, entrance width, and site layout.
Counting accuracy: Under suitable installation conditions, the HX-CCD21 can achieve a maximum people-counting accuracy of up to 99%. Actual results may vary with occlusion, crowd density, and environmental conditions. This figure should not be interpreted as the accuracy of every AI function.
Data connectivity: Check network conditions, available interfaces, and integration requirements with existing management platforms.
Privacy management: Before deployment, establish clear requirements for data collection, storage, access permissions, and compliance with applicable local regulations.
7. Frequently Asked Questions
1. What is the biggest difference between HX-CCD21 and traditional people counting systems?
The HX-CCD21 combines dual-lens 3D vision with AI analytics. In addition to basic entry and exit counting, it supports ReID deduplication, employee identification, age and gender analysis, and dwell-time measurement, giving retailers more dimensions for understanding store traffic.
2. Does HX-CCD21 use facial recognition?
Our people counting approach uses 3D vision and human feature analysis. Facial identity recognition is not required for basic traffic counting. ReID is used to associate potentially repeated human detections and is not the same as verifying a person’s real identity. Actual data handling should be checked against the deployment configuration.
3. How does ReID help retail stores?
ReID can associate human detections across different times or monitored areas. This helps reduce repeated records when the same visitor appears more than once, improving the usefulness of traffic measurements.
4. Does the HX-CCD21’s 99% accuracy apply to every AI function?
No. The up-to-99% figure refers to maximum people-counting accuracy. It does not mean that age analysis, gender analysis, employee identification, and ReID deduplication achieve the same accuracy. These functions should be evaluated separately.
5. Do small retail stores need an AI people counting system?
It depends on the store’s requirements. A basic counter may be sufficient for occasional entry measurements. If a store needs continuous traffic records, employee filtering, dwell-time analysis, or longer-term traffic trends, an AI people counting system can provide additional data for operational analysis.
Conclusion
People counting should not stop at recording how many people enter a store.
At FOORIR, we aim to help retailers gain a better understanding of customer traffic through dual-lens 3D vision, AI human detection, ReID deduplication, and employee identification.
By bringing these technologies together, our HX-CCD21 provides a broader foundation for analyzing visitor movement and store traffic, supporting decisions about store operations, workforce planning, and marketing evaluation.
The goal is to move beyond simply counting people—and toward understanding how traffic patterns shape retail operations.
