Foot traffic and occupancy.
No cloud subscription.

People counting, queue detection, zone occupancy, and flow analytics — running on the cameras you already have. For retailers managing conversion and staffing. For hospitality operators tracking occupancy and guest flow. No cloud processing. No per-device subscription fees. No biometric data collected.

Five capabilities. Built equally for retail and hospitality.

Measure foot traffic with people counting.

Track entries and exits across store entrances, hotel lobbies, restaurants, and venue access points using edge-based people detection models. Generate real-time occupancy counts, entry/exit ratios, and traffic metrics that can be integrated with analytics, workforce management, and operational systems.

Zone occupancy and dwell-time analytics.

Track real-time occupancy across floor zones. Measure dwell time per zone to identify high-engagement areas, dead zones, and traffic patterns. All processing on-device — no video leaves the location. Used for retail merchandising optimization and store-layout testing, restaurant table-turn analytics, and hospitality amenity-usage analysis.

Queue detection and wait-time estimation.

Detect queue formation, measure queue depth, and estimate wait time at checkouts, service desks, and front desks. Trigger staffing alerts when queue length or wait time crosses a threshold. Built for retail checkout optimization, food-service throughput, returns-desk staffing, and hotel front-desk and concierge queue management.

Customer flow and merchandising intelligence.

Aggregate anonymized movement patterns to generate store-wide traffic heat maps. Identify path-to-purchase routes, conversion bottlenecks, and underperforming merchandising zones. Data aggregated on-device only — movement pattern analytics exported, never individual shopper tracking. Retail-specific use cases include store-layout A/B testing, endcap performance analysis, and seasonal display optimization.

Property operations for hospitality.

Lobby, restaurant, common area, and amenity occupancy tracking across hotels, resorts, and hospitality venues. Real-time occupancy data integrated with property management systems for housekeeping dispatch, F&B staffing, amenity load balancing, and energy management. Pairs with people counting for whole-property utilization analytics.

No faces. No biometrics. No consent flows required.

ModelNova retail and hospitality models are designed for anonymous count and flow analytics — not identification. No facial recognition is performed. No biometric data is collected. Video inference runs on-device and frames are discarded after inference.

No facial recognition

Models classify presence and movement patterns, not identities. No face data is captured at any point in the pipeline.

No biometric collection

Output is anonymous counts, occupancy states, and flow metrics — never personally identifiable information.

No video transmitted

Frames are discarded after inference. Only structured count and flow data exits the device.

Compatible with consumer privacy expectations and applicable data protection regulations in most jurisdictions, without requiring biometric processing agreements or customer consent flows.

What this looks like in production.

Representative deployments, based on the silicon and models we ship today.

Vignette A — Retail

A 150-store retail chain deploys ModelNova people counting and queue detection on existing in-store cameras across all locations. Hourly footfall data feeds workforce management — staffing matches actual traffic, not a six-week-old forecast. Queue-length alerts trigger checkout-staff dispatch in real time. Average customer wait time drops 30-40% within the first quarter; cashier hours fall 10-15% without queueing complaints.

Vignette B — Hospitality

A boutique hotel group runs ModelNova occupancy sensing across lobby, restaurant, pool deck, and gym at twelve properties. Housekeeping dispatch responds to actual amenity usage, not a fixed schedule. Front-desk staffing adjusts to lobby traffic in real time. F&B sees a 15-25% improvement in staff-to-cover ratios during peak periods.

Total shopper or guest data leaving the location: zero.

Go deeper.

ModelNova Zoo

Browse pre-optimized people counting, occupancy, queue detection, and segmentation models. Free to evaluate.

Custom Model Development

Beyond off-the-shelf — bespoke models for retailer-specific use cases like SKU tracking, shelf monitoring, planogram compliance, and loss prevention.

Fusion Studio

Retrain Zoo models on your own footage. Tune to your store layout, your camera angles, your conditions — and ship a fine-tuned model in days.

Request a demo

Tell us about your business. We’ll come back within two business days with a tailored demo and a deployment cost range.

Foot traffic and occupancy. No cloud subscription.