Human-centric
AI Computer Vision
3DiVi Inc., founded in 2011, is one of the leading developers of AI and machine learning (ML) technologies for computer vision.
Your request has been successfully sent. We'll get in touch shortly.
THANK YOU!

50% Less Shoplifting in 183 Clothing Stores with 3DiVi Face Recognition

Introduction

Retail theft is a growing concern, with losses in the U.S. alone hitting $121.6 billion in 2023 and expected to reach $150 billion by 2026. Organized retail crime (ORC) and repeat offenders make matters worse—nearly 60% of shoplifters return to steal again. Yet, traditional manual prevention methods are too slow to keep up.

Facing these mounting challenges, one clothing retailer turned to the 3DiVi team to integrate face recognition into their video management system (VMS)—utilizing up to 900 cameras across 183 stores in 32 cities for real-time shoplifter detection, instant alerts, and a safer shopping environment.

Challenges

  • Camera Placement: To ensure maximum detection accuracy, the face recognition system providers should offer guidance on optimal camera placement and configuration (e.g., camera angles, lighting conditions, etc.). If repositioning cameras isn’t possible, they should give system configuration recommendations to compensate for suboptimal setups.

  • Real-Time Processing: Notifications for identified shoplifters must be delivered to store security teams instantly, within 5 seconds. The system should be able to support up to 900 cameras across 183 stores in 32 cities, handling a peak load of around 50 faces per second.

  • Minimizing False Positives: The system must be fine-tuned to handle issues like occlusions from masks, hats, or sunglasses.

  • Database Management: The system should support history storage for up to 3 months, allowing quick data retrieval for incident investigations through face-based search.

  • Integration with Existing Security Systems: Flawless integration with the existing Loss Prevention workflows, including incident notifications, investigations, and resolutions.

Solution

A joint solution combining 3DiVi technology and our partner's software was implemented to address these challenges. 3DiVi Omni Platform served as the face identification engine, while the partner's software provided a web interface for staff with role-based access to manage cameras, blacklists, and alerts.

Key Functionality:

  • Real-Time Shoplifter Detection: The system identifies known shoplifters as they enter the store and sends real-time alerts to the security officer’s mobile devices.

  • Historical Tracking: Visitor data is stored for up to 3 months, allowing fast searches for past visits using facial images.

  • City & Store Segmentation: Shoplifter lists are customizable by city and store, with real-time alerts directed to the appropriate security team at each location.

Implementation

The partner deployed Hikvision cameras with Face Capture capabilities in stores to capture faces and send cropped facial images to the central server.

The central server was used to extract biometric templates, compare them against the blacklist, and send a notification if a match was found.

This combination of local face capturing and centralized identification ensured high face recognition accuracy while maintaining low network bandwidth usage.

3DiVi Omni Platform was integrated in a straightforward 3-step process:

1️⃣ Server Software Deployment: 3DiVi Omni Platform and partner software were installed on Linux servers, with configurations for monitoring, alerting, and backup operations.

2️⃣ Camera Setup: New cameras were installed, and existing ones were reconfigured using a camera check tool 3DiVi Cam QA to optimize face recognition performance.

3️⃣ System Configuration: Roles and user permissions were set up, city-based segmentation was established, blacklists were populated with initial facial images, and real-time notification systems were activated.

Once integrated, the system operates as follows:

  • Real-Time Monitoring: The system continuously scans store visitors and cross-references them with the database in real-time.

  • Automated Alerts: Security teams get immediate alerts when a match is found, enabling them to respond quickly.

  • Data Insights & Reporting: Retailers can track incidents, review security footage, and refine their theft prevention strategies, all from one centralized platform.

Results

After integrating 3DiVi Omni Platform, the retailer experienced impressive results:

  • 50% reduction in shoplifting theft within just one year!

  • Faster Incident Resolution: With real-time suspect identification, security teams could act immediately, preventing further losses.

  • Improved Efficiency: By linking related incidents across multiple stores, security teams could reduce investigation time and increase the percentage of resolved incidents.

Conclusion

With real-time face recognition, instant alerts, and optimized incident management, our partner transformed their retail security, cutting theft by half in just one year.

Want to stay ahead of retail crime? Book a consultation with 3DiVi experts to see how our AI-powered solutions can secure your business.

Read More Success Stories

Omnigo Software
Ecortex
Mobile-Technologies Inc.
Papillon APFIS
Unique Technologies
Papillon ACS
Start your project with 3DiVi