Shulou(Shulou.com)12/24 Report--

With the rapid development of science and technology and the deepening of digital transformation, smart city has become an important trend of urban development. Especially in recent years, with the arrival of the AI craze, the city has entered a new 3.0era, which is a new chapter full of innovation and change. The rapid iterative development of smart cities is inseparable from the strong support of Intel.

On October 24, the 2023 Intel Mathematical Intelligence Park and Community Ecology Conference with the theme of "Xinqi Mathematical Intelligence to create Huicheng" was successfully held. At this conference, Intel comprehensively demonstrated its vision and strategic layout in the field of smart cities, talked freely about the future trend and development direction of smart cities, and jointly explored the latest technologies and products in the field of smart cities.

At present, with the rapid development of new technologies such as 5G and big language models, smart cities have entered a new stage of data-reality integration and intellectual symbiosis, which has brought great innovation potential and value to various industries. In order to solve the security problems faced by smart cities in the process of development and the challenges caused by road congestion and inadequate infrastructure, Intel has built a strong portfolio of edge artificial intelligence software and hardware products and vertical industry solutions. work with a wide range of ecological partners to accelerate the digital and intelligent transformation of cities.

According to reports, Intel is not only committed to building a variety of hardware product portfolios for edge computing, including Xeon processors, Core processors and ARC series, Flex series independent graphics cards, but also provides powerful development tools and AI software products, such as OpenVINO tool kits and Geti platforms. At the same time, Intel is committed to creating innovative solutions to create critical value for different business types and end customers.

For example, VPPSDK solution can help customers quickly develop NVR / VPP product applications on Intel CPU platform, so that developers can directly use the interface provided by VPPSDK to complete the corresponding functional development without in-depth understanding of Intel chip-related underlying video processing software. In addition, Intel also launched an edge AI computing box, which combines advanced software and hardware technologies and commercial AI algorithms to help industry partners accelerate the launch and deployment of solutions.

In the keynote speech at the conference, Guo Wei, general manager of Intel's Internet of things and Channel data Center in China, expressed his views on the smart city. "the purpose of wisdom is to make the city more and more warm and life more comfortable." in this process, Intel first of all does its job, and the chip has to do a good job, from Core, Lexus, to strong, to GPU, IPGA, WIFI, 5G, etc., to make a good connection. Do a good job of computing as the basic hardware of a smart city to help the development of the city.

At the conference, Intel demonstrated the advanced solutions for the future smart city created by many partners.

For example, based on Intel processors, Gao Xinxing Technology Group built a holographic intelligent video application scheme. With the characteristics of global video AR, video service scene and application function diversification, this scheme can help to realize the holographic processing of intelligent video and multi-dimensional data. By creating a "video + AI+AR + big data" fusion platform, Gao Xinxing can provide video storage and transfer, intelligent parsing, AR tagging, analysis and decision-making and other capabilities to support intelligent and holographic business applications.

"the main purpose of Gao Xinxing is to create an AR+AI intelligent holographic intelligent video solution. I hope our technology can open another window for the industry. We can build all kinds of linkages better on the basis of ordinary surveillance and ordinary intelligence. We can't do all kinds of information applications and development without computing power, and Intel can provide basic computing power." Chen Lijun, chief technology officer of Gaoxing Technology Group, concluded.

Entropy-based technology, as one of the partners, also demonstrated the entropy-based intelligent edge server BioCVBox built jointly with Intel at this conference.

This is a lightweight application server product tailored for small and medium-sized projects. BioCVBox is based on Intel ®Core ™i3 processor, pre-installed 64-bit Window10 Enterprise Edition operating system, and equipped with Entropy-based Wanao Ruida intelligent integrated management platform. It integrates storage, computing, management and other applications, and has the characteristics of low construction cost, simple deployment, operation and maintenance, strong performance, security, stability, high reliability and so on.

Hu Xuehui, general manager of Entropy Technology China Marketing Department, said that in smart cities, the needs of customers will become more complicated, and their own performance, scenarios, and business needs will also become diversified. We hope to have high-quality partners like Intel to give us a blessing. Next, entropy-based technology will focus on the joint development of the ability to add generative AI edge servers. In the future, the cooperation between the two sides will be extended to the three ends of the cloud, and the overall collaboration will enable more smart city scenarios such as parks and communities.

In addition, there is an open-domain digital store solution based on Intel ®edge AI computing box as the core. Based on Intel ®ARC series independent graphics cards, Shenzhen Yuanji Technology Co., Ltd. has created a multi-channel GPU server.

Intel through cooperation with partners to provide a variety of linkages to achieve information, digital management, improve efficiency and reduce costs. Intel believes that collaboration with partners is a virtuous circle, which can enable Intel to allocate high-quality resources in this direction in a more targeted manner, which will ultimately benefit society.

Conclusion as the world's leading semiconductor chip manufacturer and technology innovator, Intel actively lays out the smart city field. Through its strong hardware product portfolio, AI software product and solution innovation capabilities, Intel and ecological partners work together to promote the digital intelligence transformation of smart cities. In this process, Intel's technology empowerment will bring more opportunities and value to the development of smart cities.

In the future, Intel will continue to join hands with more partners to promote the construction and development of smart cities and create a better living environment for people.

After the conference, CTOnews.com and other media also conducted in-depth dialogue and exchanges with Intel, Gao Xinxing Technology, etc., and discussed many possibilities of making the city more "smart".

Here are some shorthand notes of the conversation.

1. Compared with the PC and automotive application scenarios on the consumer side, what is the difference in the hardware requirements of smart cities?

Renu N. Navale (Vice President and General Manager, Smart City Business Unit, Network and Edge Solutions, Intel):

Compared with on-board and other traditional PC applications, smart city and smart community classification usage scenarios are quite different. First of all, it is diversified, there are various forms, we have various forms in the scope of the city. So we have video analysis, including reasoning for AI data, and based on this scenario, we combine Intel CPU and all kinds of hardware, including D / GPU just mentioned, as well as our software like OpenVINO, into a video edge box and a more powerful video edge server, and merge it into different scenarios, in urban scenarios. There will be, for example, wide temperature requirements, shockproof requirements, anti-corrosion requirements and so on, these are diversified, customized requirements, these are the requirements of some specific scenarios. At the same time, we also need to maintain the optimization of the overall product system. These are what Intel strives to pursue, in addition to in-depth optimization, but also to cooperate with partners to achieve the extreme for market segments.

Guo Wei: this problem is in two aspects, that is, "the same" and "different". The same places, such as PC and cars, all have the requirements of calculation, and the requirements for the park also keep pace with the times. This is the same in smart communities and smart cities.

The difference is that, in addition to diversification and variety of product forms, in many new technologies, smart parks and smart communities will be faster than all kinds of PC and cars, for example, large models and multi-modes, and it is easier to form commercial closed loops in smart parks and smart communities.

2. Intel released the smart community solution in November last year. What new progress can Intel share in the smart community solution over the past year?

Renu N. Navale: a lot has happened in this year since November last year. The application of real artificial intelligence began to break out. So you learned about the big model that includes the whole ChatGPT language. In addition to language, we also have the explosion of large video models, and we also have a variety of perceptual fusion, with the help of a holographic system, which is a major direction we see. Intel has spent a lot of time on this this year, working with our partners to really land it.

At the same time, 5G construction has been very mature, and now it is really large-scale integration of 5G with high computing power and high reliability to provide a truly sustainable market. And we want to do one thing, but also for sustainable development, so there is green energy and efficient treatment, these are our technological efforts.

Guo Wei: from last year to this year, we are in the smart community, smart park, Intel as the basic software and hardware supplier, last year we thought more about how to use our technology to support customer management, as well as smart park, smart community management, how to support our customer service, or to enable our customers to better serve their customers.

The whole real estate industry may slowly move from the secondary industry to the tertiary industry. How Intel provides good hardware and software to support it. This is a question we have considered over the past year.

3. With the rise of the big model, whether some changes have taken place in the core and shape of the wisdom park. When we do specific landing construction, what are the obvious differences?

Guo Wei: I think from the perspective of the business department, the big model provides a very important technical means. Because whether it is a large model or a generative AI, it is a technology. From the wisdom park, the smart community, the core of the smart city or the needs of the city itself, I think it has not changed. We still want to live a warmer, more comfortable life and have a stronger predictability of all kinds of disasters.

In fact, we look at the traditional AI technology, is a lot of unstructured data into structured data, using a large model, so that these structured data can better see the meaning. If the large model comes out, it can solve the hidden meaning of unstructured data which can not be solved in traditional AI, and more predictive analysis can be made. This has a lot of guidance for our wisdom park and wisdom community. I think the change in the kernel can make it possible for us to do what we should have done, or what we want to do but can't do.

Renu N. Navale: technically, we can see that traditional AI may be trained with a large amount of data so that more accurate data can be obtained after training. But after the implementation of the large model, because with the enhancement of computing power and different processing capabilities, we can achieve very high accuracy with a small amount of data, and we can provide, compared to our solution, there is a retraining mechanism at the edge, which can provide very good customization capabilities.

For example, in the community, we may have many competitors, such as education and community service providers may be different, different from place to place. The factory is the same, for example, on the production line, maybe the products I produce are different, and the final accuracy requirements are not the same. We do a retraining in the operation through the big model, which can be very accurate. This point was also mentioned by our partners in the speech.

Hu Xuehui: the media teacher may see an oven-like device in the Demo display area. In fact, it is an edge server made by us in conjunction with Intel, which empowers us and provides generative AI technology. At present, there are some pilot applications of generative AI technology in some intelligent city segmentation scenarios overseas. For example, we are in an Indonesian hospital, and we have a guidance system at their nurse station. By implanting our technology into the agent robot, it will be convenient for nurses to perform a good voice assistant function through this robot when the patient or the patient's family is busy. Fall into the wisdom park business, there are also better application scenarios. For example, visitor guide, virtual receptionist. In addition, because we also have some deep experience in the field of intelligent office, we will also follow up the application of AI attendance integrated with digital people.

Chen Lijun: the big model is quite popular in recent years, more is ChatGPT or Baidu "Wen Xin one word", more is the common language category. In fact, this is universal, for all kinds of industries, such as subways or financial things. As far as equipment is concerned, it is not accurate to put it alone in an industry. It should be a small society, and it is an application of some achievements of various industries, whether it is the AI computing power or the big model computing power, in this small society of the park.

4. In the digital wave, how does Intel, which has been ploughing the Internet of things for many years, view the industry prospects and market demand?

Guo Wei: I have been doing the Internet of things for the past few years, that is, selling Intel's CPU to various industries. We obviously feel two points.

First, the whole industry has broad prospects for development. Just like the smart city, the smart city is actually a place that accommodates all kinds of industries, and now you see more and more places that need computers, as well as in various industries in the country. We are the first to do more in the industry is partial to retail, retail has gradually developed into intelligent transportation and education, and in recent years we have started industrial, medical and intelligent transportation. In fact, more and more industries need computing nodes, and more and more computing nodes are needed, which itself is an opportunity for Intel.

Second, with the rapid development of China in recent years, the proportion of the tertiary industry is getting higher and higher, that is to say, the proportion of the service industry is getting higher and higher, which requires more and more IT to support. Originally, it was supported by people, but now it is supported by computing, which is also the result of the development of this industry. Therefore, in terms of the prospects of the whole industry, whether it is Intel itself or, more importantly, the development of various industries in China, I think there are more and more opportunities behind.

5. From the perspective of the Chinese market, the current trend of intelligent city construction in China is obvious. Based on the current strategy, from which dimensions will Intel expand its business layout?

Guo Wei: first, we should do our job well and provide more and more high-performance and diversified chips.

These chips may be in addition to CPU, including display chips, including a variety of programmable chips, including connection chips, and so on.

Second, the software level. We need to provide a variety of software tools, such as the OpenVINO just mentioned, as well as a variety of middleware, as well as a variety of reference software tools that can support and serve customers.

Third, the customer level. In fact, Intel has been deeply ploughing various industries in the past few years, establishing deep cooperative relations with the core partners of each industry, as well as the most important partners in each industry, so that they can support and serve the partners in the industry well, so that these partners can finally serve the development of the industry as a whole.

I think it is about these three aspects that we can continue to enter new industries. As Intel technical experts just said, we may not have thought about it ten years ago, but looking at the development trend of the whole of China in the past few years, whether it is intelligent pension, smart medical care, or intelligent industry, China is developing rapidly from the low end to the middle and high end. This is the further development of the country, and our industry layout continues to follow the development of the country and the development of society.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.