We all surely know the advanced analysis systems for athletes’ movements that have been used for some time to optimize their performance. These systems, based on AI, try to obtain the maximum possible from each athlete. But that is not the only area where artificial intelligence can make contributions to the sports sector…
Beyond training
Obviously, for athletes, having a machine that suggests improvements in the way you train or move is a great advantage. However, today’s artificial intelligence finds application in many other areas, just as important. From Marketing, through journalism, to medical diagnoses, arbitrations and the best sports equipment, you name it.
And one of the areas that greatly benefits from the tremendous analytics capacity that AI offers, is video analysis of sporting events. In major championships, hundreds or thousands of hours of video are generated containing all types of sports, people, events, records or interesting moments. The problem is that watching all this material manually, in order to index it and comment on it, is simply impossible. For this reason, video analytics systems, such as Videoma Archive, can help manage this huge amount of data in a reasonable amount of time and, not only that, extract all the possible information, automatically.
Use cases for video analytics in sports
There are a whole series of cases in which automated video analysis can provide interesting or necessary information to multiple processes, within various areas, in numerous sports. These are some of them, although there are many more:
- Find appearances of a certain athlete. Whether it is locating one’s own athlete or someone specific within a team, Videoma Archive’s facial biometric systems are capable of analyzing dozens or hundreds of hours of video using facial recognition, to find all the moments in which the athlete you search appears on the screen. This allows for all the clips to be gathered automatically, in order to review them, comment on styles or techniques, or simply to create a clipping book of all the interventions of said athlete over time.
- Locate certain sports. In mass broadcasts, such as international athletics championships or even the Olympic Games, there are dozens of television channels, and not all of them prioritize the same sports in the same way. Thanks to AI and some special analyzers, it is possible to select certain sports on all of those channels and mark the time points (or automatically extract clips) when they appear on the screen. This way it is not necessary to schedule recordings of multiple channels and, in addition, you will never miss anything, thanks to permanent monitoring.
- Locate statements. Speech-to-text (S2T) transcription systems allow you to locate certain phrases or words in a matter of seconds. With its help we have the possibility of locating and saving statements that have been made and that interest us, both our own and those of others. On the other hand, there is the possibility of finding specific terms such as “goal”, “out” or similar, to go directly to the points of interest of a specific match or competition.
- Locate certain sounds. Similarly, there are analyzers capable of finding certain sounds (a starting shot, audience shouts, etc.) within video recordings or broadcasts, so that we can gather all those events in a single report and view them when we need it.
- Visual learning. 65% of people learn better visually than auditory. This means that having an automatic selection of their interventions in multiple competitions makes it possible to better analyze what they have done well or badly, and helps to improve performance so as not to make the same mistakes again. In fact, this same thing can be applied to studying the competition, and watching their tactics or techniques.
Automatic reports
Videoma Archive not only allows us to analyze thousands of hours of video unattended in 24/7 to locate all the information we need, but also creates a searchable database, which will help us find more data and relate information to each other.
The platform has a web interface that allows easy access from almost any device, and generates automatic reports summarizing everything it has found, for an easier and more targeted analysis.
Conclusion
If you need to manage all the recordings of your athletes on one platform, have the ability to search for certain details in a matter of seconds, among thousands of hours of video, Videoma Archive is the platform that makes it easy for you. Its AI analysis capacity can be expanded to cover all your needs.
If you want to know more about Videoma Archive, you can visit us at booth CS302 at ISE 2024 (Integrated System Europe) to be held at Fira de Barcelona, from January 31 to February 2, 2024.