CES 2019: Intel, Alibaba Group Create AI-Powered 3D Athlete Tracking Tech for 2020 Tokyo Olympics

Intel and Alibaba announced on the eve of CES that the companies are teaming to develop athlete tracking technology powered by artificial intelligence (AI) that is aimed to be deployed at the Olympic Games 2020 and beyond. The technology uses existing and upcoming Intel hardware and Alibaba cloud computing technology to power a cutting-edge deep learning application that extracts 3D forms of athletes in training or competition.

“Optimized to run on Alibaba Cloud infrastructure, Intel developed a novel technology concept called 3D Athlete Tracking that uses AI to analyze video of athletes in ways that are expected to improve both the training process and has the potential to enhance the audience experience during actual competition,” says Navin Shenoy, executive vice president and general manager, Data Center Group, Intel. “This technology has incredible potential as an athlete training tool and is expected to be a game-changer for the way fans experience the Games, creating an entirely new way for broadcasters to analyze, dissect, and reexamine highlights during instant replays.”

How It Works: The combination of computer vision with AI deep learning algorithms will generate a 3D Mesh that enables coaches and trainers to extract intricate real-time biomechanical data — via multiple standard video cameras without the use of special sensors or suits — from which they can analyze performance and introduce new training enhancements.

Specifically, the technology leverages advanced pose modeling techniques and other AI algorithms designed to analyze the biomechanics of an athlete’s movements. The performance is captured with regular video cameras, the AI algorithm is applied with a heavy dose of computing power and a digital model of the performance is created that can be analyzed in different ways.

Intel and Alibaba, together with partners, aim to deliver the first AI-powered 3D athlete tracking during the Olympic Games Tokyo 2020. The partnership seeks to provide athletes with new training data and analysis and provide fans insight into how world-class athletes perform and compare against one another.

“We are proud to partner with Intel on the first-ever AI-powered 3D athlete tracking technology where Alibaba contributes its best-in-class cloud computing capability and algorithmic design,” says Chris Tung, CMO, Alibaba Group. “With our technological advancements, Alibaba is transforming the sports media and broadcasting industries. We continue to collaborate with industry leaders and drive innovations to create new experiences for the world to enjoy.”

How Intel Participates in Singles Day: The Olympic endeavor is the latest collaboration between Intel and Alibaba. Recently, Alibaba deployed Intel’s latest processor and memory technology to help power the 11.11 Alibaba Global Shopping Festival. The festival set a record $30.8 billion of gross merchandise volume in transactions within 24 hours, and has exploded past both Black Friday and Cyber Monday in popularity and sales.

The Alibaba Global Shopping Festival places a lot of demands on Alibaba Cloud’s infrastructure to perform real-time processing of massive spikes of data and to support heavy amounts of traffic visiting its website. To meet the computing demand, Alibaba deployed next-generation Intel Xeon Scalable processors along with Intel Optane DC persistent memory in its production environment. This new Intel platform can efficiently process massive amounts of data in real-time, enabling Alibaba’s digital commerce applications to deliver a smooth, responsive user experience.

What Alibaba Used: Alibaba’s highly interactive and data-intensive applications require the infrastructure to keep large amounts of hot accessible data in the memory cache to achieve the desired throughput (queries per second) and deliver a smooth and responsive user experiences, especially during peak hours of the Alibaba Global Shopping Festival. The large capacity of Intel Optane DC persistent memory allowed more data to be stored closer to the processor, allowing for significantly higher infrastructure efficiency and lower total cost of ownership.