This article originally appeared as part of the industry’s ONLY full CES show coverage in the February 2024 issue of Security Business magazine. Don’t forget to mention Security Business magazine on LinkedIn and @SecBusinessMag on Twitter if you share it.
LAS VEGAS – Searching CES for video surveillance innovations is a difficult prospect. It is quite the feat to wade through the dozens of DIY cameras, baby monitors, and doorbell cameras to uncover tech that may apply to enterprise and professionally served residential markets. There were a few companies who fit the bill at CES 2024:
Rajant: Bringing AI to the Edge for Video Analytics
I met with Steven Griggs of Rajant Corp., who demoed the company’s new “Cowbell” – a computing hub that the company is touting as “the future of AI deployment.” It is designed to run AI applications directly on end-devices.
“This is an AI and a machine learning platform,” Griggs says. “You can teach this device anything and run your own video analytics application – the possibilities are really endless. We are demoing a camera and video analytics for applications like license plate recognition and facial recognition. You can teach it what a gun looks like, or what violent action looks like.”
Thanks to embedded wireless and Ethernet connectivity, the devices are essentially transporting AI over the air and into any devices that can be networked to them – enabling mobility and more. “If you need more computational power, all you need to do is turn on another one, and it shares the load and distributes the computational power,” Griggs says.
Hailo: Edge-Based Video Processing
Yaniv Iarovici of Hailo demonstrated the recently launched Hailo-15 processor chips, designed to be integrated into intelligent cameras to deliver video processing and analytics at the edge. In security applications, this enables faster and more accurate detection of people, objects, and situations, at a high resolution, even in challenging lighting conditions.
Iarovici showed an integration with a VMS and NVR that aggregated video feeds at the network edge, with Hailo chips applying AI analytics to them. CVEDIA’s object and people detection application was running on 30 streams.
The company also demonstrated the power of AI-enabled video enhancement. Hailo-based edge AI dramatically improves image quality and provides a clear, sharp image – even from poor-quality, unstable, or low-light video.
Swann: AI for Video Doorbells
Swann HomeShield is a generative AI-powered video doorbell assistant designed to automatically answer when someone rings the doorbell. Product manager Steven Davies walked me through several automated reply scenarios, including package delivery, solicitations, and even emergency situations.
Digital Barriers: Hi-res Video over Cellular
In the AT&T booth, Digital Barriers demoed an AI-based video codec that adjusts the bitrate of streaming video for network optimization.
“We are taking the feed from a standard IP camera, reducing the bitrate to about 600 kbps, and streaming it to our server in Plano (from Las Vegas), and back again, with about half a second delay,” Sales Engineer Dustin Jones explained during a demo.
Not only does the system reduce the amount of bits being transmitted, it also actively adapts to network changes and traffic. Jones says the company will be demoing the system at ISC West.
Paul Rothman is Editor-in-Chief of Security Business magazine. Email him your comments and questions at [email protected]. Access the current issue, full archives, and apply for a free subscription at www.securitybusinessmag.com.
