FaceFirst is a recent entrant providing real-time suspect alerting using surveillance video of faces.
The system is designed to use megapixel cameras and tracking servers at the facility being monitored. The tracking servers connect to FaceFirst's monitoring center where matching of suspects to watchlists occurs. FaceFirst will provide staff to visually verify and confirm alerts. Those alerts can be forwarded to appropriate personnel/authorities. View their architecture for an overview of this approach. Also, note the solution is based on Cognitec's facial recognition algorithms.
Outside of the most critical infrastructure, we are skeptical of the technical success and financial viability of this approach. For background, note our concerns about the accuracy claims and operational issues involved in facial recognition.
On the positive side, FaceFirst's approach integrates monitoring staff to handle false alerts. This will relieve the organization's security personnel who are generally not staffed to handle this.
On the negative side, for almost all users, this will require the deployment of new megapixel cameras and new servers on site. Adding this capital cost to the ongoing operational expense of the monitoring staff will make this an expensive proposition.
Compare to Dragnet Solutions who has proposed a similar architecture/model for facial recognition. Contrast to 3VR, who offers facial recognition bundled into a mainstream DVR/NVR.
Verint, which is best known in the physical security industry for video surveillance but has built a sizeable cybersecurity business as well, was...
The world's leading video surveillance information source, IPVM provides the best reporting, testing and training for 10,000+ members globally. Dedicated to independent and objective information, we uniquely refuse any and all advertisements, sponsorship and consulting from manufacturers.