The International Olympic Committee Awarded SwiMMinD for Using Emerging Technology to Make Video Analysis Accessible to Beginner Swimmers
Today our lives are intertwined with digital media and it is difficult to define how long a person should be in front of a computer or mobile phone screen. Childhood obesity and diabetes continue to grow in alarming rates. Research reveals that children consume between 3 to 6 hours of media in a typical day. This includes using computers, cell phones and tablets for games, music and reading. Because of that, children tend to exercise less! That is why SwiMMinD sprang in the autumn of 2017 with a goal to connect the digital world with the world of swimming by solving the following problems:
• Time-consuming Video Analysis for beginner and elite level swimmers;
• Childhood obesity and diabetes;
• High costs to race internationally;
• Majority of Learn to Swim Coaches struggle to follow Learn to Swim Programs or to provide certificates to their beginner swimmers.
SwiMMinD’s team includes highly skilled and experienced professionals ranging from Artificial Intelligence (AI) Experts, Educators, Developers, Olympic Champions, World Record Holders and Learn to Swim Experts. There is no other competitor in the video analysis sector that has this unique combination of focused expertise and experience.
Thomas Lurz, an Olympic and World Championship medallist, a legend in marathon swimming has been with SwiMMinD since its inception. He joined SwiMMinD because “it focuses on beginner swimmers.” He is certain that if beginner swimmers receive more feedback “they will be able to progress much faster,” especially when learning happens in a fun, interactive way.
Lurz continues in saying that “the idea to have children swim more via mobile application is AMAZING! Parents will be enabled to check with their mobile phones the progress of their kids, SwiMMinD will award their kinds and allow them to compete globally.” Perfect symbiosis between digital and active life!
Another reason why SwiMMinD focuses on beginner swimmers is because traditional video analysis companies focus on elite level swimmer. Those companies provide video analysis tools that are very-time consuming. In other words, it is difficult for coaches to provide video analysis to all their swimmers because it takes hours to make the analysis with their drawing tools. In this respect, SwiMMinD thinks that faster analysis is much better. Therefore, SwiMMinD aims to develop automatic video analysis to beginner swimmers which shall be based on established coaching standards, Learn to Swim Programs.
Overall, SwiMMinD is an AI Assistant in a mobile app which shall instantly deliver meaningful video analysis based on well-established Learn to Swim Programs. It shall provide interactive certificates and the Characters of the Learn to Swim Levels and Certificates (Seahorse, Torpedo, Goldfish, Shark etc) shell communicate with its users. Health and Fun!
The AI Assistant, SwiMMinD shall also provide video-based competitions to RACE with swimmers around the world. In this respect, the World Champ, Thomas Lurz highlighted rhetorically “Could you imagine the smile of a kid from Würzburg who could compete with a kid from New York without leaving their local pools?”
Lida Tsonkova, Founder of SwiMMinD with Learn to Swim experience in America, Australia, Bulgaria and Spain states that in her career as a swimmer (multiple National medallist in Bulgaria and Spain) and as a coach she came across video analysis, but those tools were made only for highly experienced professionals in the video analysis or Coaches who spent countless hours deciphering those tools. At the end, the present market offers labour-intensive video analysis tools which are not helpful to beginner swimmers.
“We keep asking ourselves? What excites children? What motivates a 9-year-old swimmer? What would a 50-year-old beginner swimmer needs or wants? How could parents help the process of learning to swim of their child without interrupting the swimming lessons? How could we excite more parents, grandparents and children about swimming? How could we bring more attention to beginner swimmers? How could beginner swimmers learn faster?” While answering those questions Lida goes through different versions of the app on a daily basis in order to make it better.”
The technology behind SwiMMinD is Borislav Popov (Bobi) and Georgi Dimitroff (Goro). Bobi was a Head of Semantic Analytics at Ontotext where he built a team that developed a Dynamic Semantic Publishing platform used by leading media, scientific and financial institutions in Europe and the US. He specialized in Natural Language Processing and his primary interests are the possibilities appearing from automatically linking unstructured information with semantically modeled structured data, as well as the resulting multi-paradigm indices and higher order analytics.
Georgi Dimitroff has a PhD in mathematics from TU berlin and specialized in stochastics and later in Semantics and AI. He was behind the innovations in the proprietary machine learning libraries and personalized engine of Ontotext and is behind the applied AI solutions for Euromoney, BBC, FT, IET, S&P Global and others. His primary R&D interest is in the area of neural nets and deep learning. Goro is the key figure in solving the primary AI water environment challenge which is a core intellectual property of SwiMMinD.
“After over 15 years of working on how to make it easier for broadcasting corporations such as BBC to discuss sport events, it is time for me to get involved in the noble cause to make children more sports oriented … more active …get excitement from both the digital and the sports world” To achieve this, Bobi stated that SwiMMinD shall distinguish body parts and movements by accurately compiling images from videos. The event stream processing shall enable real-time image recognition using deep learning models. Bobi concluded that “By combining our deep learning models, we could train our models in memory in the cloud.”
SwiMMinD models on body movements in the water using machine learning algorithms, which by nature improve on performing a task as they gain more experience. Thus, SwiMMinD shall automatically assign a value to each action, such as a back float, a starfish in the water. A perfectly executed starfish, for example, has a high value, but a starfish with bended arms or legs would have a lower value.
Valentin Milkov, the Co-Founder of SwiMMinD, US Open Champion and a World Cup Director says that the artificial intelligence and machine learning will play an important role in the swimming analytics in general. “Existing mathematical models develop existing knowledge and insights in swimming, while AI will make it possible to discover new connections that would not be obvious for coaches, swimmers or parents. Thus, SwiMMinD will not only help people Learn to Swim based on well-established international standards but also SwiMMinD will offer the optimum swimming execution of a stroke”
As a consequence of all of the above, just few months after its inception, in January 2018 SwiMMinD won “Most Business Value” Smarter Project Award from the International Olympic Committee and Alibaba Cloud. Due to its innovation, SwiMMinD was exhibited at the winter in the 2018 Olympic games in Pyeongchang, South Korea.