Vivotek unveils edge computing facial recognition camera
A trio of edge AI developers have revealed new technologies that bring biometric storage and processing close to the place of application. They include a new camera from Vivotek, a family of vision-processing chips from Hailo, and a real-time data processing platform from BrainChip.
Vivotek has launched its first facial recognition camera, FD9387-FR-v2, that combines edge computing to identify gender and age from video footage even when people wear masks. The company says it can store up to 10,000 profiles with a 99 percent accuracy rate and is compliant with the U.S. National Defense Authorization Act.
The FD9387-FR-v2 Facial Recognition Camera from Vivotek integrates the SAFR Inside AI facial recognition platform from RealNetworks, Inc.
Features include real-time facial detection and tracking, early warning of strange faces, image privacy mode for sensitive areas, strong cybersecurity protection with encrypted data storage and transmission.
The Vivotek FD9387-FR-v2 Facial Recognition Camera suits building banks’, retailers, and buildings’ automation/access control systems. For example, it can integrate with business intelligence (BI) services to send real-time notifications when VIP customers enter the store.
Additionally, it helps track traffic in and out of smart buildings, adding an extra layer of security. Unauthorized visitors can be reported and recorded for future reference.
Hailo introduces new vision processor for edge-based smart cameras
Hailo, a chipmaker for edge AI processors, has released its Hailo-15 family of vision processors. These high-performance chips can be integrated into cameras to provide advanced video processing and analytics at the edge.
According to the company, Hailo-15 offers deep learning video processing and AI performance, allowing it to be used by city operators, manufacturers, retailers and transportation authorities for various applications. It can help improve safety and security, increase productivity and machine uptime, protect supply chains, and detect incidents quickly.
“With this launch, we are leveraging our leadership in edge solutions, which are already deployed by hundreds of customers worldwide; the maturity of our AI technology; and our comprehensive software suite to enable high-performance AI in a camera form factor,” states Orr Danon, the CEO of Hailo.
The Hailo-15 VPU family has three variants (Hailo-15H, Hailo-15M, and Hailo-15L) to meet different processing requirements. With performance up to 20 TOPS, the company says it enables over 5x higher performance than existing solutions at a comparable price point.
According to the company, this AI-based video analytic solution provides enhanced safety, privacy and cost-efficiency to organizations while reducing the complexity of network infrastructure.
Hailo also claims that its low-power, fanless processors are well-suited to industrial and outdoor applications where dirt or dust may otherwise impact reliability.
“With a single software stack for all our product families, camera designers, application developers, and integrators can now benefit from an easy and cost-effective deployment supporting more AI, more video analytics, higher accuracy, and faster inference time, exactly where they’re needed,” adds Danon.
Hailo will demonstrate its Hailo-15 AI vision processor at ISC-West in Las Vegas, Nevada, from March 28-31.
The company began a collaboration with Innovatrics on facial recognition late last year.
BrainChip unveils second-generation Akida platform
BrainChip Holdings Ltd. has announced the second generation of its Akida platform. It drives hyper-efficient and intelligent edge AIoT devices with advanced capabilities such as 8-bit processing, time domain convolutions and vision transformer acceleration.
“We see an increasing demand for real-time, on-device intelligence in AI applications powered by our MCUs and the need to make sensors smarter for industrial and IoT devices,” says Roger Wendelken, the senior vice president in Renesas’ IoT and Infrastructure Business Unit.
Applications that can benefit from fast vision transformation include facial recognition and other biometrics.
Akida’s second-generation technology features Temporal Event Based Neural Nets (TENN) spatial-temporal convolutions, which can efficiently process raw time-continuous streaming data. The technology enables more straightforward implementations while providing high accuracy and a lower development cost and is suitable for use in industrial, automotive, digital health, smart home, and smart city applications.
Akida’s second generation also offers Vision Transformers (ViT) acceleration. This neural network excels in computer vision tasks such as object detection and semantic segmentation, BrainChip says. Akida’s ability to process multiple layers simultaneously and hardware support for skip connections enables it to self-manage complex networks like RESNET-50 entirely within the neural processor without CPU intervention and reduces system load.
“With the addition of advanced temporal convolution and vision transformers, we can see how low-power MCUs can revolutionize vision, perception, and predictive applications in a wide variety of markets like industrial and consumer IoT and personalized healthcare, just to name a few,” says Wendelken.