The latter is aimed primarily at enterprise access control applications, with a capability of recognizing personnel and support for 3D liveness detection. What’s more, its interface is contactless, helping to ensure that workplace conditions are sanitary in the wake of the pandemic.
The OnPoint reader benefits from the upgraded Vision AI software, which has been refined to bring more efficient computing from servers to edge devices while maintaining accuracy. In announcing the upgrade, Oosto explained that Vision AI uses Kubernetes technologies to ‘containerize’ Oosto’s computer vision software, allowing its neural networks to run on edge devices.
The neural networks themselves, meanwhile, have been optimized to support low-power edge devices, with Oosto asserting that its algorithms are seven times more efficient, in terms of watts-per-video stream, than those of its closest rival.
Dieter Joecker, Oosto’s Chief Technology Officer, explained that not only do Vision AI’s new enhancements allow computing workloads to be moved from expensive servers to edge devices, “but, perhaps more significantly, we’re doing this without sacrificing recognition accuracy, security and performance in real-world scenarios or expanding the physical footprint of the system.”
The upgrade comes after Oosto’s announcement of a major security deployment at Australia’s prestigious Australian Turf Club horse racing venue this past summer.
October 25, 2022 – by Alex Perala