World cups, league tournaments, Grand Slams and Super Bowl! It’s safe to say that the buzz around sports is never-ending. However, the prevailing situation is a challenging run for sports and fans, alike. There has been a steep slump in viewership over the last couple of months. ESPN’s viewership has dropped by 50%, while the NBA has seen a 14% decline in viewership contrasted to 2019.
The current pandemic has affected many sections of the sports media due to social distancing measures and government-imposed restrictions. Nonetheless, sports fans across the world seem to be eagerly awaiting their next game. A study conducted by Forbes about sports fans showed that though there has been a lack of live games, fans are as hopeful as ever. For instance, the Green Bay Packers that compete in the NFL have a season-ticket waiting list with 137,000 people.
This brings up the inquiry – how can sporting leagues and broadcasters improvise to keep fans at the centre and what role can technology play when it comes to fan engagement?
How AI can help in maximizing the value of existing sports content archives.
Blending Sports with Storytelling
Storytelling in sports is becoming a new trend. OTT platforms are featuring documentaries, created from game content and from hours-and-hours of behind-the-scenes and locker-room footage.ESPN’s docuseries featuring Michael Jordan and the legacy of Chicago Bulls (The Last Dance) has seized audiences around the world, with over 5.6 million views since its release. Many such documentaries have released on different sports and on different OTT platforms.
This trend has made sporting leagues realize the latent potential of existing sports archives. Technology that can help to find special story-worthy moments and help in repurposing this content can provide a huge lift to fan engagement actions.
AI can help in searching for specific moments from this based on facial and emotional recognition and help discover other edit-worthy moments to fast-track the process of story creation.
Match Facts and Insights
Recently, the German Football League (DFL) announced a partnership with Amazon Web Services (AWS) to drive advanced analytics across its platform and broadcasting. In every match, DFL collects around 3.6 million data assets to gain a more in-depth insight into the game. These insights are then shared with national and international broadcasters so that their 500 million fans worldwide can understand the match, players’ positions, and the probability of goals more quickly.
Quick and Custom Highlights
During most sporting events, multiple live feeds capture the same match from different angles. And the content producers are left to manually analyze and extract relevant content from this bulk data. With the help of technology and Artificial Intelligence (AI), content creators and distributors can curate sports content with the utmost ease. Applications include the creation of:
- Quick highlights: The use of AI-enabled editing workflows can minimize the turnaround of generating stresses.
- Custom highlights: Use of AI can also create custom highlights for different user segments depending on the fanbase, interest areas, region, etc., allowing broadcasters to improve customer targeting, ratings, and revenue.
How AI gets customized for a particular sport
While techniques of tagging and analysis of sports media have been employed by various AI entities, the combined use of these techniques to create custom highlights is a novel venture. Sports media is a unique playing field for AI technology, given the abundance of information created during each event. Streamlining the analysis of this data and applying it in a financially viable manner is potentially game-changing.
To unleash the full potential of sports archives, the content needs to be enriched with high-quality granular metadata.
Hypertagging engines that have a specialized taxonomy for a particular sport leads to higher accuracy in results. This aids in better search and retrieval of specific moments from many hours of content.
The critical thing to note here is that cookie-cutter computer vision technology does not serve the needs of sports, as every game is different.
The AI developed should be contextualized with specific elements of that sport, and then the solution should be able to deliver on the last-mile journey in production environments.
What kind of information can be extracted from sports?
Meta-tagging a key moment in a game involves collating information on five important aspects.
- Camera angle detection: Information regarding whether the frame is shot from an aerial view, as a closeup, or at a field angle.
- Custom location detection: Information on the primary subject of the frame is captured. In this instance, the focus is on the key players. In other cases, it could be the crowd, the ball, or the field.
- Logo detection: The logos on the players’ jerseys or on the ground are detected depending on the camera’s angle.
- Face recognition: Most importantly, AI can also collect information on players whenever their faces are captured at suitable angles.
- Emotion detection: Detecting useful emotion sequences of players with editorial objectives in mind.
By integrating the detection of multiple facets of information ranging from camera angle detection to face recognition, sports content becomes ready to be edited. The final step is to intelligently assemble these moments to either create multiple highlights of a single game or curate numerous moments into a story which inspires.
To know more on how AthenasOwl helps sporting leagues and broadcasters in for fast searching and retrieval of content, explore our Smart Catalog solution.