Intel’s Vision for the Future of Smart Video
What is sight without the power to understand what’s being seen? As the Internet of Things (loT) revolution continues, smart video technology keeps creating huge amounts of data. But even as these video technologies become more prevalent in cities, industrial facilities, retail stores, and even private homes, it’s still a major challenge to store and analyze that data. Intel answers this challenge with a scalable end-to-end (E2E) solution for secure collection, storage, and analysis of vast volumes of loT video data.
Smarter Industries and Smart Home
Can a better world exist without the proper resources to manage the exponential increase in 1/0?
Smart cameras are critical for making cities safer and more secure. Depending on the size of the city, this metadata can stack up into terabytes— and it all has to be analyzed in order to derive useful results. Within smart cities, smart transportation systems are generating tremendous amounts of data about vehicles, passengers and roads—often creating a full terabyte that needs to be securely collected, processed and stored every single day.
In a wide range of manufacturing sectors, virtualization technology is helping raise efficiency while lowering the costs of doing business. Smart factories that use this technology can generate a terabyte of data every day. In smart retail, video generated by digital security and surveillance systems helps keep stores secure and efficient—but with more than 75 million video surveillance cameras being sold to retailers in 2018, demand enormous storage capacity.
The loT transformation doesn’t end at the commercial space. The smart home revolution is already well underway. By 2020, the global market for smart home technology is projected to reach 100 billion, and most households will have more than 50 connected devices in the home. As smart homes become commonplace, the amount of data generated by smart devices will grow exponentially—as will the storage and analysis systems required to deal with that data.
Fragmentation is a developer’s nightmare. That’s why Intel believes the only solution, is a holistic E2E approach.
As more industries adopt IP cameras, Intel’s integrated; hardware accelerators deliver the performance needed to process high-definition media in real time. These solutions ensure secure transmission and accurate analysis of data, with the help of a unique architecture in which connectivity, analytics and performance can all expand as needed.
Intel’s “develop first, port fast” toolset powers Intel’s Computer Vision software development kit (SDK). This SDK enables easy video processing, whether for in-store footage or handling security and surveillance applications. All Intel Xeon processors come ready to work with an entire portfolio of specialized libraries, saving developers significant time.
In the loT, security is at a premium. As many security experts know, hackers have already penetrated the ordinary layers of software security, and are now focusing their attacks on connected devices like IP cameras and network video recorders (NVRs). Intel’s entire architecture is designed from the ground up to defend against common forms of attack.
A Future with Intel
Intel technologies are making video data more accessible, analyzable, manageable and actionable for an entire ecosystem of developers. Each one of Intel’s video security and surveillance solutions delivers the performance required to process bandwidth-intensive video at the fastest rates possible.
Intel RealSense cameras are the “eyes” of this intelligent visual system. From handheld devices to snap-on PC cameras and beyond, RealSense makes 3D scanning and interior mapping simple, with imaging-and-feature tracking that make it easy to capture 3D models of real-world objects and rooms.
Movidius MyriadX vision processors (VPUs) are the “visual cortex” of the system, delivering top shelf performance with a very low power thermal footprint. In fact, MyriadX processors offer the best performance per watt of power of any processor under 1.5 watts—a full five times the capacity of Myriad 2 processors. These efficiencies allow more cameras to be installed, increasing in the amount of data captured, stored, and analyzed.
Finally, Intel CPUs act as the system’s “brain,” delivering best-in-class processing power packed into an ultra-thin and lightweight chip. Intel Xeon processors, for example, help deliver real-time analytics, processing for mission-critical tasks and big data insights.
Intel’s diverse portfolio of cameras, vision processors and CPUs provide the tools to make massive amounts of data easy to access, manage, analyze and transform into actionable insights.