news

한어Русский языкEnglishFrançaisIndonesianSanskrit日本語DeutschPortuguêsΕλληνικάespañolItalianoSuomalainenLatina

On July 27, Beijing time, the opening ceremony of the 33rd Summer Olympic Games was held in Paris, France, and the world entered the "Olympic time". Beijing News Shell Finance reporter found that along with the Chinese Olympic athletes shining in Paris, there are alsocloud computing, artificial intelligence and other advanced Chinese technologies.

Among them, Alibaba Cloud "took over" two-thirds of the remote distribution of live broadcast signals through its cloud computing technology, while SenseTime, Baidu, and Lenovo provided auxiliary training for the Chinese basketball team, diving team, and sailing team respectively with their own exclusive AI technology. This is in line with the statement made by IOC President Bach in April this year: With the rapid development of artificial intelligence, the Olympic movement has once again reached a crossroads where it must either actively seek change or passively change. The IOC hopes to use the huge potential of artificial intelligence in a responsible way.

Cloud computing accounts for two-thirds of the total, surpassing satellite as the main method of Olympic broadcasting

In the past, live broadcast signals of the Olympics were mainly transmitted via satellite and traditional optical cables. These facilities are not only very expensive, but also require the deployment of hardware in advance, and broadcasters cannot change their plans at the last minute. To address this problem, the Olympic Broadcasting Service (OBS), an official organization of the IOC, and Alibaba Cloud jointly developed and launched the "Olympic Broadcast Cloud (OBSCloud)", which provides cloud broadcast support for global broadcasters, allowing them to start global transmission later and more quickly. This technology was used for the first time at the Tokyo Olympics, and is a major technological advancement since the satellite TV broadcast of the Olympics began in 1964.

At this year's Paris Olympics, cloud broadcasting will officially surpass satellite broadcasting and become the main way to broadcast Olympic events to billions of viewers around the world.

Yiannis Exarchos, CEO of Olympic Broadcasting Services

According to reports, OBS Live Cloud will become the main method for remote distribution of live broadcast signals for the 2024 Paris Olympic Games. Currently, two-thirds of the remote services booked are through cloud computing. The Olympic Live Cloud will be based on Alibaba Cloud’s global deployment ofPublic CloudWith the help of the live broadcast infrastructure, the Paris Olympics is expected to produce 11,000 hours of live broadcast footage, which will be distributed from Paris to the world in this way, transmitted to more than 200 countries and regions around the world and reach billions of viewers.

“Cloud broadcasting is extremely cost-effective in achieving large-scale global broadcasting of content. The global coverage of cloud broadcasting is comparable to that of satellite broadcasting, and sometimes even higher, while broadcasters can more easily access all the content we provide. Satellite broadcasting can only support a limited number of channels, while cloud broadcasting can transmit all live content,” said Sotiris Salamouris, chief technology officer of Olympic Broadcasting Services, in an interview.

Sotiris said that the cooperation between the two parties has also launched a number of latest AI-based applications. "One of the most successful applications is the multi-camera replay system, which is a novel and effective way to produce content. We use the 'time and space freeze' technology to freeze the continuous actions in time sequence into 3D space, which not only allows the audience to see the actions clearly, but also enables 360-degree multi-angle viewing, replaying and disassembling the wonderful moments of the athletes."

AI-assisted training helps Chinese Olympic athletes win gold and silver medals

The Paris Olympics is also the first Olympics to use AI technology on a large scale, realizing the full-process and full-scenario application of AI technology. AI is fully integrated into pre-match training, in-match penalties, and can even be used to detect doping and protect athletes from cyberbullying. For example, the International Olympic Committee has developed an active AI protection tool specifically for monitoring and deleting abusive posts to ensure that athletes can compete without distraction and are not affected by rumors and slander.

Even before the competition, Chinese Olympic athletes have already used AI technology for training assistance. Among them, the Chinese national basketball team uses SenseTime's AI technology. The sports performance analysis platform for the Chinese national basketball team developed by it can analyze the movement status of each athlete and the movement trajectory of the basketball in real time by building a three-dimensional model of the venue and a 3D motion capture algorithm, helping to improve the training efficiency of athletes and provide a scientific basis for competitive strategies. This product will participate in the entire competition journey of the Chinese national basketball team during the Olympic Games, providing sports data analysis, competitive strategy optimization support, etc.

The Chinese national diving team uses Baidu's AI technology. The AI-assisted training system developed by Baidu for the Chinese national diving team is based on Baidu's big model technology such as Wenxin Yiyan. It can understand and execute complex instructions from coaches and athletes, provide accurate information in a timely manner, and can also score diving movements in real time, conduct precise quantitative analysis, and provide guidance. Previously, Olympic champion Quan Hongchan said that she would use the AI ​​diving-assisted training system to pick out the details of the movements during her training.

The Chinese National Sailing Team uses Lenovo's AI technology. During the training phase, the personal intelligent agent "Xiaotian" built into Lenovo AI PC can assist athletes in formulating exclusive training plans. It can also use AI computing power to output analysis of body movements and movement trajectories from videos shot during training, helping the team to grasp a more comprehensive training situation.

In addition, the entire event of table tennis and archery in this Olympic Games adopted SenseTime's smart event broadcasting technology. This technology can achieve multi-person, large-scale, and multi-angle scene coverage using 3D non-perceptual motion capture technology without any wearable devices, and obtain spatial motion posture information in real time. It can present the landing point, rotation speed and direction of the table tennis ball, and the flight trajectory of the arrow to the audience in front of the screen through AR.

Through AI technology, direction visualization, motion trajectory description, ball landing point display, etc. are displayed.

In addition, many domestic large model apps have also launched Olympic-related functions, such as the Tongyi APP, which has launched functions such as Event Encyclopedia, National Cloud Sports, and AI Sports Photo. SenseTime's large model provides ordinary audiences with functions such as the "second-by-second painting and fun shooting" templates that can generate Olympic-style photos.

According to estimates by market research firm Mordor Intelligence, the market size of AI+sports will be approximately US$6 billion in 2024. By 2029, the size of this market is expected to grow to US$21 billion, with a compound annual growth rate of approximately 28.69%.

"Artificial intelligence can provide more athletes with personalized training methods, better sports equipment, and more personalized fitness and health plans. Artificial intelligence will greatly improve the efficiency of organizing sports events, change the way sports are broadcast, and make the audience experience more personalized and immersive," said Bach.

Luo Yidan, financial reporter of Beijing News Shell

Edited by Bai Huabing Proofread by Zhao Lin