A shopping mall in Bangkok · Retail · shopping mall, stores and events
120+ cameras and 20+ AI capabilities in one mall system
A shopping mall in Bangkok uses Nara to turn its cameras into one system for both retail analytics and security: visitor counts, demographics, engagement and zone traffic, alongside intrusion, loss prevention and safety alerts. Everything is processed on-site and shown on dashboards, and the data is shared with other systems over MQTT.
- cameras on one Nara deployment
- 120+cameras on one Nara deployment
- AI capabilities deployed
- 20+AI capabilities deployed
- edge devices running in sync
- 4edge devices running in sync
- Customer
- A shopping mall in Bangkok
- Sector
- Retail · shopping mall, stores and events
- Deployment
- On-premise: Advantech hardware, mall cameras; data shared with other systems over MQTT
- Partners
- Advantech
Challenge
Mall operators need to know how many visitors they get, who they are and where they go, to manage tenants, marketing and events. They also need constant security coverage across a large public space. Separate single-purpose systems for each need are costly and do not share data.
Solution
One Nara deployment provides more than 20 capabilities. Analytics: people counting in and out with person re-identification so no visitor is counted twice, zone traffic, pet and pet-cart counting, anonymous age and gender, uniform detection to separate staff from visitors, shopping-bag analytics, expression analysis, and engagement (time in each zone and where people look). Security and safety: asset-removal alerts, watch-list person detection, intrusion, height-risk areas and climbing on railings, fire and smoke, weapons and fights with tracking of the people involved, falls and people lying down, and abandoned bags.
How it works
- 01inputMall cameras
- 02processPeople detection, tracking and ReID, feeding analytics and security models
- 03actionDashboards, alerts, and MQTT to other systems
Results
- 120+ cameras on a single Nara deployment
- 20+ AI capabilities, with several models running on each camera
- 4 edge devices running in sync, sharing data
- 100% processed on-site, in real time
Nabrio components
- IP multi-camera nodes
- People detection
- Tracker with line and zone counting, dwell time, heatmap and abandoned-object detection
- Person re-identification and uniform matching
- Face detection and recognition
- Age and gender, facial expression, head pose and gaze
- Pose and action recognition (fall, fight)
- Fire and smoke, weapon and object detection
- MQTT publish
- Dashboards
Use cases
Build the same thing on your site.
cameranara seesPeople counting
Simple door counters double-count people who return or use several entrances.
Heatmaps and dwell time
Operators can count visitors, but cannot see where they go, how long they stay or what catches their attention.
cameranara seesVisitor demographics
Operators know how many people visit, but not who they are, which makes tenant, marketing and event planning guesswork.
cameranara seesStaff presence monitoring
Staffing must match demand and standards, but managers lack a real-time view across stores.
Intrusion detection
Sites need to know straight away when people or vehicles enter closed areas after hours, without a guard watching every screen.
See what Nara sees, on your own cameras.
Download Nara and build your first flow in minutes, or tell us what you need to see and do, and we’ll show you how it runs on your own hardware.