Video Analytics

Often, yes. Many modern IP cameras ship with built-in analytics, and server- or cloud-based analytics can process streams from cameras that don’t. We assess your existing system during design and tell you plainly which cameras can carry analytics, which need a firmware path and which are worth replacing.

Both models are in active use. Edge analytics run on the camera itself, which scales without server hardware and suits distributed sites; server- and cloud-based analytics suit deeper processing and mixed camera fleets. Most of the systems we design combine both, matched to each site’s network and camera system.

Traditional motion detection triggers on any pixel change — headlights, rain, animals, shadows. Analytics classify the cause of movement before raising an alarm, so responses are reserved for people and vehicles that matter. For monitored sites, that means fewer guard dispatches and less after-hours disruption.

Video analytics is the working application — counting, detection, plate reading. Deep learning is the technology many modern analytics are built on. If you want the engineering detail, our Deep Learning Video Surveillance page covers how the models classify, search and detect. Deep Learning Video Surveillance

Analytics that process footage of identifiable people fall under the Privacy Act 1988 (Cth) in Australia and the Privacy Act 2020 in New Zealand, alongside state-based surveillance legislation. Obligations vary by analytics and context, so we design data retention, signage, and access controls into the system rather than leaving them as an afterthought. This is general guidance, not legal advice.

Rarely. Analytics from AXIS, Avigilon, Genetec and AerVision integrate with the major VMS platforms we deploy, and edge analytics are largely VMS-agnostic. Where a platform genuinely limits what analytics can achieve, we’ll show you the gap and the options — including ones that don’t involve starting again.