Intel recently appeared at the EdgeX+OpenVINO Ecological Partner Conference , and explored the future of intelligent edge together with edge computing and visual reasoning industry experts from innovative companies such as Alibaba Cloud, Lenovo, VMware, Thundersoft, EMQ, and Dao Keyun.
“
In the past year, we have witnessed the rapid development of EdgeX and OpenVINO™. More and more developers and ecological partners have joined the intelligent edge ecosystem, and through innovative solutions and technologies, they are constantly promoting the iterative update of edge computing and artificial intelligence.
-- Dr. Zhang Yu
Chief Technology Officer, Network and Edge Division, Intel China
Intel Senior Principal AI Engineer
”
EdgeX+OpenVINO™ Ecological Partner Conference is an important event in the field of edge computing and artificial intelligence. The rapid development of edge computing technology provides a powerful boost to the realization of intelligent edge applications, and OpenVINO™, as an industry-leading visual reasoning tool suite, provides powerful algorithm support for efficient visual analysis and intelligent edge applications. Since the first release of the OpenVINO™ toolkit in 2018, Intel has paid close attention to market demand, focused on future development trends, and continued iterative updates. Recently, it released the OpenVINO™ 2023.0 version, which not only gives it higher performance, but also makes it easier to use. Flexible, open and comprehensive, with a view to helping AI developers simplify workflow and improve deployment efficiency to a greater extent.
With the vigorous development of edge computing, using artificial intelligence technology to deploy innovative application solutions on the edge side to change the traditional business form has become a new choice for many modern enterprises. In order to help ecosystem partners complete solution construction and deployment more efficiently and conveniently, Intel not only provides a series of advanced software and hardware products and technologies such as Intel® Core™ processors and OpenVINO™ tool suites, but also uses Intel® Video AI Computing Box (hereinafter referred to as AI Box), adopts an integrated design that integrates software, hardware and algorithms to accelerate the development and engineering implementation of edge AI applications. AI Box integrates Intel's mature software and hardware platforms, and can help it efficiently build various edge AI solutions with high performance, low power consumption and optimized for edge AI loads by integrating rich AI algorithms of ecological partners. At the hardware level, AI Box is based on the new Intel Core processor and a unified SoC hardware platform, equipped with Intel Iris Xe integrated graphics card, which can provide high-performance media capabilities and AI reasoning capabilities for edge video analysis at low power consumption, greatly improving single Thread and multi-thread performance, support high-density streaming and integration of various workloads, applicable to various restricted embedded environments, and can effectively reduce solution costs and deployment and operation costs. At the software level, AI Box integrates various software development tools such as Intel Media SDK, OpenVINO™ tool suite, SVET tool suite, System Studio, OpenCV, etc. It can use Intel® FML library for feature matching, optimize video processing models, and accelerate AI reasoning etc. to provide multi-dimensional support. With a reference architecture that integrates video decoding and analysis functions in a computing box, and end-to-end configuration of applications with flexible AI capabilities and video analysis pipeline reference implementations, AI Box can help developers greatly accelerate the development process.
With the support of advanced software and hardware, Intel AI Box has been accelerating the implementation of intelligent edge applications in various industries around the three levels of computing power, algorithms, and ecological construction. Based on this, in order to effectively deal with the challenges of edge AI fragmentation and accelerate solutions to the market, Intel also combined AI Box and EdgeX to make its applications not only focus on the field of intelligent video analysis, but also use a more powerful ecosystem to mine More intelligent edge application scenarios such as camera management, sensor fusion, and cloud-edge collaboration to help more industry ecosystem partners and end users in retail, industry, and medical care quickly realize the construction of edge AI solutions and product verification, thereby effectively lowering the development threshold , greatly speeding up the time to market. For example, in the medical field, with the help of AI Box, EdgeXFoundry can help medical institutions conduct AI-assisted diagnosis and remote consultation. By connecting various medical sensors and cameras, and distributing data and video streams to artificial intelligence algorithms and remote terminals, EdgeXFoundry can help R&D teams accelerate R&D progress and access various types of equipment, allowing R&D teams to focus on business processes. In the future, with the advantages of EdgeXFoundry's edge-side platform, the technical architecture can be smoothly migrated between various medical devices.
In the era of the Internet of Things, terminal applications are blooming everywhere, bringing about the growth of massive data. The combination of artificial intelligence and edge computing can effectively alleviate the data flood caused by the vigorous development of the Internet of Things. In the future, Intel will continue to help industry ecosystem partners explore the vast opportunities in the edge AI market through comprehensive product technology leadership, innovative solution driving force, and ecosystem building capabilities, and empower thousands of industries to move towards a more digitally intelligent future.
©Intel Corporation, Intel, the Intel logo, and other Intel marks are trademarks of Intel Corporation or its affiliates. Other names and brands involved in this article are the property of their respective owners.