Standard tier captures what people do — anonymous, no consent needed, works anywhere. Deep tier captures what people feel — emotion signals, requires delegate opt-in. Both deliver the same core output: a timestamped engagement timeline your team can act on.
No facial analysis. No biometrics. No consent required. Works at any event for any audience.
FACS Action Unit processing. VAD modelling. Requires delegate opt-in via event registration.
Four steps from sensor to report. No facial analysis at any point. Equivalent to an advanced people counter with engagement intelligence built in.
Sensors placed at zones you want to measure. No specialist hardware required. Works with standard venue infrastructure.
Footfall counted, dwell time measured, movement vs stillness tracked, crowd density mapped. No facial analysis at any stage.
Zone engagement scores, footfall curves, and dwell time data delivered live. Accessible on any device during the event.
Post-event report with engagement score over time, peak moments, drop-off points, zone heatmaps, and audience size curve. Delivered within 5 days.
No facial recognition. No biometrics. No individual data of any kind. Standard CCTV signage noting that analytical sensors are in use is sufficient in most GDPR contexts. No consent forms. No data subject rights to administer. Full documentation →
Five steps from camera to emotional intelligence report. Facial muscle movements are analysed and discarded immediately — no faces, no biometrics, no identity information stored at any point.
Standard cameras placed at zones. Delegate opt-in collected at event registration before the event opens.
44 FACS Action Units extracted from each frame — the discrete facial muscle movements behind every human expression.
Raw frame deleted immediately on the edge device. Only numerical AU values and derived emotion scores leave the device.
Engagement, confusion, delight, and scepticism scores updated every 3 seconds by zone. Confusion alerts fire when thresholds are crossed.
Engagement timeline, emotion breakdown, peak moments matched to content, confusion hotspots, staff effectiveness, and recommendations.
Delegate opt-in required — typically one checkbox at event registration. No faces or biometric templates stored. Aggregated signal scores only. We provide DPIA template, consent language, Legitimate Interests Assessment, and Data Processing Agreement. Not facial recognition. Not EU AI Act high-risk. Full documentation →
This matters for procurement, legal sign-off, and EU AI Act compliance. Standard doesn't touch faces at all. Deep analyses muscle movements — not identity — and is not classified as high-risk AI.
No facial analysis at any point. Behavioural signals only. Functionally equivalent to an advanced people counter.
Analyses muscle movements, not identity. No face images stored. No biometric templates. Not EU AI Act high-risk. Not Article 9 GDPR.