Article by
Max Kalmykov
Senior Vice President, Media & Entertainment / New York, USA
In recent years, there has been a notable rise in the investment by streaming platforms in acquiring live sports rights globally. Jacob Feldman, a Sports Business Reporter at Sportico, explored some of big players that have entered the live sports rights landscape recently: Netflix announced a deal to stream two NFL games on Christmas Day next season; Apple committed to a 10-year partnership with Major League Soccer; Google is spending $2 billion annually for NFL Sunday Ticket rights, and this list goes on.
With the potential that advertising can bring to streaming platforms and technological advancements that allow tackling the live streaming latency issues, there's at least one more essential area that generates revenue streams and requires solid technology — live sports analytics.
In this article, we'll take a closer look at the technological aspects that are essential for the smooth and efficient live streaming of sports analytics collection, processing, and display.
Sports Analytics Powering Sports Streaming
Live streaming sports analytics involves a structured review of recorded footage to assess player performance, team interactions, and the effectiveness of tactics used during a match. This method includes recording games or practice sessions and utilizing specialized software to analyze the visuals, often in slow motion or frame-by-frame, allowing for insights that may not be apparent during live action. Modern sports video analytics is based on technological advancements such as slow-motion playback, multiple-angle views, and data integration.
Sports analytics serves multiple purposes, varying based on the needs of its users. Broadcasters use analytics to evaluate team performance and identify historical trends, which enhances their commentary and improves viewer engagement. Investors look to data to assess the value of sports club sponsorships, helping them make smart decisions based on audience reach and brand visibility. For sports bettors, analytics provides valuable insights that can predict event outcomes and shape effective betting strategies. By examining historical data and player stats, bettors can make more informed choices. Teams also use analytics to boost fan engagement and improve merchandising strategies, tailoring their marketing efforts to better meet fan preferences and increase merchandise sales. Overall, sports data analytics is reshaping broadcasting, investment, betting, and fan interactions in the industry.
Essential Technology Aspects for Live Video Sports Analytics
Modern sports analytics technology can help enhance performance, engagement, and decision-making. These advancements range from real-time video processing and object detection algorithms to the integration of sensor data and machine learning for predictive analytics. Cloud-based storage solutions and customizable dashboards also further guide teams and broadcasters by facilitating efficient data management and visualization.
Real-Time Video Processing
Real-time video processing usually involves the use of high-performance GPUs and specialized software frameworks (such as OpenCV and TensorFlow) to rapidly process video frames. Methods like frame differencing and optical flow are employed to detect changes and events as they happen. This technology facilitates immediate feedback and insights during live events, enabling real-time analytics for player performance, game strategies, and fan engagement data. These real-time insights can improve broadcast quality and deliver dynamic content to viewers, including instant replays and statistical overlays.
Object Detection and Tracking Algorithms
Object detection and tracking related to live sport events analytics involve employing algorithms to recognize and monitor players and objects, such as the ball, within video feeds. These algorithms are divided into single-object tracking, which targets one entity, and multiple-object tracking, which follows several objects at once. Techniques used in this area include convolutional neural networks and sophisticated tracking methods like Kalman filters or SORT (Simple Online and Realtime Tracking). These approaches analyze each frame to identify and classify objects while keeping track of their identities over time. The advantages include accurate player tracking, improved game analysis, and enhanced fan experiences through augmented reality applications.
One more approach to ensure effective object detection and tracking is using LIDAR — a remote sensing method for determining ranges (distance, height, etc.) and creating accurate 3D maps and models. In 2022, DataArt partnered with iReality, a company that builds its technology on LIDAR and provides next-generation sensing, tracking, timing and data augmentation solutions for horse racing. The primary use of LIDARs was to capture the landscape around a specific point. iReality approached DataArt to find out if it was possible to accurately track moving objects with LIDARs and build a final solution if the test showed promising results. DataArt proved the technical potential of LIDAR usage for moving object real-time tracking by using a system of LIDARs, an additional camera, and building a workable tracking algorithm.
The new technology:
- Could work with any number of LIDARs and cameras and included an administration management panel and incident-resolving system
- Captured data from sensors sent to cloud
- Processed data (image recognition and tracking algorithms)
- Streamed data – giving detailed information about each horse’s position
Learn more about cutting-edge solutions for live sports analytics, including object detection and tracking using AI, LIDAR technology, and sensor data integration. Discover how machine learning, cloud-based video processing, and customizable dashboards enhance game analysis, player performance insights, and fan engagement while driving innovation in sports video analytics.
Top comments (0)