AI and Machine Learning(AI/ML) have gotten much attention in the last 5 years. AI and Convoluted Neural Networks (CNN) seem well suited for extracting finer detail from video images. The process is basically one of taking a set of inputs and mapping them to a set of outputs. But unlike a direct mapping that might occur with linear regression techniques, CNN mapping includes several hidden layers that result in non-linear transformations of input to output. As Gary Marcus of New York University points out in a critical appraisal of deep learning, the methods for achieving that mapping are complex, with results often not understood by the team creating the network. While beyond the scope of this post, for those wanting more insight into CNN and Machine Learning, see Google’s developer section for an introduction to feature identification.
Training a CNN requires that large data sets be run through the fitting routines. In the case of tracking human motion, one needs i) large numbers of images with the relevant features identified, ii) images need to include people in various poses consistent with the activities that are to be tracked, and iii) subjects and backgrounds need to be sufficiently diverse that the models can track in all environments.
Validating the trained CNN is equally daunting. Typically, a model trained on say, 90% of the dataset is used to test its ability to accurately identify the remaining 10%. Randomly picking the 90% in repeated tests allows for a very large validation test.
With images of the features taken from multiple digital videos located in different viewpoints, standard DLT techniques can be used to locate those features in 3D space.
As mentioned, the shape/silhouette approach offers fast processing and reasonable accuracy of gross motion. But its methods are limited in terms of the resolution that can be achieved.
AI/ML on the other hand, enable extracting the position of individual features such as eyes, joint centers and condyles. But the processing of video files is time consuming and not yet real time requiring several minutes to process videos of short duration. The computational requirements are also more expensive involving high end graphics cards and parallel processing on the GPU.
The Future
Our first experience with markerless motion tracking used shape/silhouette technology. However, our current belief is that AI and CNN are especially suited to tracking human motion and are, in fact, the future of motion tracking. Its primary shortfall is the time it takes to process video data, and that will be addressed as computing power and programming techniques evolve.
With AI/ML many of the traditional biomechanical analyses remain in place. For example, rigid body analyses where 3 non-colinear markers are used to track the orientation of a body segment find parallels with markerless tracking. Feature identification that includes a proximal joint center, lateral condyle and distal joint center could be right out of traditional biomechanical marker sets.
The high resolution of MI/ML also enables monitoring of such things as how a ball is held during a pitching activity.
AI/ML approaches are being commercialized by several companies including Theia, KinaTrax, Intel 3Dat, and SwRI. While we are actively evaluating all, Theia and KinaTrax have already demonstrated acceptance in their respective markets.
Thus far we have created tight integrations with Theia and KinaTrax. It is important from an “ease of use” perspective that the steps of recording video, processing video, applying AI modeling, and generating the appropriate analytical output progress without intervention after clicking the record button. The MotionMonitor’s existing structure is being used to collect digital video synchronously with standard laboratory peripherals such as forceplates and EMG. This greatly simplifies and expands the usability of Theia and Kinatrax markerless tracking.
In addition, The MotionMonitor’s unique design supports fast special purpose application development on multiple platforms. The In Game Baseball application developed for KinaTrax shown in the following video is a good example of this capability.
Happy smiling mature man and old woman cycling on a swimming bike in swimming pool. Happy and healthy senior people enjoying swimming with young woman. Fitness class doing aqua aerobics on exercise bikes in a swimming pool. Photo Credit: iStock
Coaches and athletes look for every legal edge over their competition. However, you may be putting too much emphasis on new innovations. Sleep is a foundation of human performance. It has the potential to fuel or wreck an athlete’s performance. It may not be flashy, but your athletes may not be getting the sleep they need to have an edge on and off the field.
Counting Sheep Is An Athletic Advantage
Why do the Boston Red Sox have a “sleep room” in their club house at Fenway Park? Why do the Seattle Seahawks wear Readibands to monitor player’s sleep? And why do MLB starting pitchers travel ahead of the rest of the team on road trips?
“Sleep is probably the most important thing we talk to [players] about,” said Brad Pearson, head athletic trainer for the Boston Red Sox. “It’s critical to their recovery. You pick the low-hanging fruit first, the simple things. You create a stable base and build on that.”
If your athletes aren’t getting the right amount of sleep, it will dilute every other aspect of practice, preparation, and recovery.
Proper Sleep Gives Your Players an Edge
Sports science is clear that an athlete’s performance is impacted by the quantity, quality, and timing of their sleep. As a coach, it’s important to understand the impact that sleep (or lack thereof) is having on your players.
1. Risk of Injury
A recent study found that adolescent athletes who get less than 8 hours of sleep per night increase their risk of injury 1.7x compared to those who get at least 8 hours.
Injuries can decimate your roster and your shot at a winning season. If you can’t keep players on the field, you begin every game at a disadvantage.
2. Physical Effects
Studies reveal that inadequate sleep is linked to a reduction in a player’s natural growth hormone, slowed reaction times, and a weakened immune system.
This means that athletes who don’t get good sleep don’t get the full benefits from your strength and conditioning program. They won’t be as big, fast, or strong as competitors who sleep well.
3. Cognitive Effects
Sleep impacts your athlete’s brains, specifically the executive functions of the brain. These functions include decision making, concentration, memory, information processing, planning, and emotion regulation.
This carries a wide variety of on and off-field consequences. For example, athletes who don’t get adequate sleep make more mental mistakes and lose their composure more readily. They are at a greater risk for depression and don’t perform as well in the classroom.
A great example of an elite athlete who understands the cognitive (and physiological) benefits sleep offers, and benefits greatly from it, is five-time Super Bowl champion quarterback Tom Brady. He seeks every advantage to keep his mind and body operating at peak form, including going to bed at 8:30 PM every night.
4. Performance Effects
By increasing the amount of sleep they were getting, basketball players improved shooting percentages by 9%. The same was true of serve percentages in tennis players.
Athletes who don’t get adequate sleep are more likely to miss shots, drop catches, strike out, and get out hustled by opponents.
According to Sam Ramsden, the Seahawks’ Director of Player Health and Performance, “Some of the best players on the team are the best sleepers.”
How much sleep do your athletes need?
According to the National Sleep Foundation, the assumed 8 hours per night may not be enough for your athletes.
The Sports Biometrics Conference is about strength and performance training, sports science and the research key organizations have been conducting about this discipline. Neuroscience is a key element as well.
Sponsor or attend?
Attending the conference gives you two days of excellent education and networking opportunities to meet like-minded professionals from across the globe. There is a registration fee as this is a professional conference. Sponsoring is for companies looking to meet these industry professionals and talk about solutions offered. Sponsorships are tiered based on levels of participation and engagement sought.
How is the event going to run during Covid-19 restrictions?
We have been planning to run the event live in San Francisco as it has for the past five years but the restrictions in California on live events over 50 people will most likely force us to convert to a fully online program. The online conference will still run on a Pacific time zone clock both days and still include all of the conference sessions, keynotes, presentations and exhibitor booths.
Can I arrange a group registration for my coaching team?
Yes, we do have group registration discounts. Please contact the conference manager at [email protected] to discuss your specific request.
Does the conference issue CEUs?
The conference has issued CEUs every year and will be working with several associations to have them credential in March 2021. Our conference program usually needs to be fully designed with all speakers in place for us to be able to apply for the credentials.
We are a start-up company. Can we still sponsor?
Our programs are designed for teams of all kinds and for companies at all stages of development to take part. We have unique opportunities for start-ups and emerging companies. Please contact the event manager.
Can I substitute someone if I can no longer make the conference?
Yes, substitutions are allowed. Please contact the event manager and we will take care of the name swap online. The event archives will enable you to watch conference session recaps following the event if you are not able to take part live during the event.
This Privacy Policy covers information which you give us, Synapse Events LLC and the Sports Biometrics Conference (“SBC” or “we”), about yourself and your professional background when you register for or attend our face-to-face events in the United States, visit or participate in any of our online programming including web sites, online courses or webinars. We maintain this data on servers hosted in the United States and possibly elsewhere. A link to the privacy policy will be visible on this website at all times.
Our business is about education on sports science, use of biometric monitoring and neuroscience in and around professional and college level sporting issues and connecting buyers and sellers of products and services in this marketplace together face-to-face and online. From time to time we work with other companies to build and link web sites, jointly produce webinars, sponsored webinars and other programming that allows you to connect directly with those companies. Where possible, we include a link to the other company’s Privacy Policy. Our Policy will govern our collection and use of information about you. In any case, we recommend that you find and read the other company’s policy as well to learn how that company will treat the information about you.
Information That Identifies You
We rely on the information you provide to us as you register, contact us, participate in surveys, post comments to public areas and interact with our web sites, respond to our email campaigns, and participate in live events, webinars or online events. We may also add data from external sources including your professional interests on websites other than ours. We will add information about how you interact with our offerings, your subscriptions, the events you attend, the content you download or read, and the like.
We may work to expand our communities by acquiring names and contact details from other sources that have compiled contact information. This information might be public information or data you provided with necessary permissions. We may also invite you to join our communities via third party social and professional networking sites.
Online Data Collection – Technologies
We use cookies and common data collection technologies, including web visitor software tools that create cookies on our web pages, in newsletters, and in our email campaigns to collect information both on an aggregate and individual basis.
Some of our web sites use Adobe Flash to provide video content, and the Adobe Flash Player has the ability to separately store information about that particular Flash application. We do not associate this Adobe Flash stored information with you as an individual. Please see the Adobe Flash Player Help page for additional information about controlling the settings of the Adobe Flash Player on your computer. Similarly, we use YouTube and Vimeo for video serving and Google, YouTube and Vimeo may issue tracking cookies. These are not passed to the Sports Biometrics Conference nor do we store them.
We gather date, time, certain information about a user’s browser and system or device configuration and capabilities, navigation history and IP address for all visitors to our web sites. We use this information for our internal security audit log, for trend analysis and system administration and to gather broad information about our audiences and their geographic locations. Among the tools deployed for this is Google Analytics, with no personally-identifiable data captured only the click path an individual might take through the event website.
How We Use Information About You
We use information our data collection has about you to provide the services and event notifications. Please help us keep your registration information up to date. You may update most online registrations by visiting the event registration page associated with a specific conference or webinar.
Information We Share About You
We share information about you when a participating exhibitor or sponsor at one of our events uses an in-booth lead capture or conference session room monitor scans your attendee credential. Information in this case is limited to your primary contact information and never includes sensitive financial data related to your registration.
When participating in a webinar or certain face-to-face conferences we may give our partner or sponsoring organization(s) access to your registration data less any sensitive financial transaction information as part of an event attendance list.
When attending our face-to-face events, it is up to you to decide whether to leave your business card or allow your badge to be read by a particular exhibitor. Badges cannot be read remotely. If you allow your badge to be read by a particular exhibitor, we may share certain information about you. A sponsor of a special, sponsored conference session will also be given information about you if you choose to attend that company’s session. Your future interaction with online content from an exhibitor or sponsor to whom you provided your contact data to may also be tracked and disclosed to that exhibitor or sponsor.
Registering online for one of our face-to-face events using an exhibitor-specific registration code that you received from that exhibitor may trigger a data share of all such exhibitor-specific registration code users. In this case we may share certain non-sensitive information about you with that exhibitor.
We may also disclose information about you to other third parties, including, for example, law enforcement or other entities in order to respond to court orders or valid subpoenas, or to protect human safety or our networks or property, and third parties as part of a merger or sale of some or all of our business assets.
You May Opt Out of Communications
You may opt out of our email marketing programs and lists by following the link at the bottom of each of our promotional emails. Please note that we may have created segmented lists tied to our different live and online events and you may be opting out selectively. To remove a specific email address from all Sports Biometrics Conference files and databases please send the request to [email protected].
Please note that you may still receive email messages from us about other companies’ offerings as part of a list rental or partner marketing programs – but only if you have given us permission to do so. These names would include past attendees to a specific live or online event where the email marketing opt out does not remove you from having been an actual customer. Similarly, past purchases cannot be removed but we do place you on our universal opt out list that is suppressing opt out names before each and every outbound marketing campaign. Most of our email marketing takes place through MailChimp and we adhere to their very strict opt-out and unsubscribe policies. Some past event data resides on our Constant Contact email marketing account where we also adhere to strict opt-out and unsubscribe policies.
Cookies On Our Site
We place cookies that identify you or allow third parties to do so on our behalf. These cookies help identify you as a returning website visitor enabling you to return to a page or offering you previously viewed.
Ability to Correct Data
We will correct any data we have about you; simply email us to request this.
Contact Us With Questions
For questions about this Privacy Policy, please contact: [email protected]
After spending a lifetime in the sport, in 2015 I realized I had many questions about swimming for which very few answers were available.
How do you explain feel for the water to a swimmer who doesn’t already have it?
What is the difference between a distance swimmer and a sprinter?
Which is more important—pulling or kicking?
What happens after a few hundred meters, when the initial great feeling wears off?
Looking at the big picture, I thought that myths, beliefs, and traditions weren’t allowing proper development of swimmers, and I was working half-blind trying to improve the athlete’s technique using trial and error. Moreover, the performance of many swimmers was getting worse while I knew for a fact that they were in better shape than ever before.
To help the swimmers progress, I needed to answer the very basic questions that had bothered me for a long time. Therefore, I decided to build a sensor that measures the propulsive force of the hand strokes.
Feel for the Water?
Feel for the water—or hold on the water—are common phrases referring to the swimmer’s ability to gain traction. While it seems to be a natural ability with gifted swimmers, it also can be taught with various drills (at least to some extent). The “feel” is vitally important to swimming success. However, it’s very difficult to explain and impossible to compare without quantifying it somehow. Fortunately, if you can feel the hold, you can also measure it.
A good feel for the water should efficiently generate propulsive force to the swimmer. Essentially, you should know the magnitude, direction, and timing of the force. As examples, the force of the strokes for a distance swimmer and a sprinter are presented in Images 1a and 1b. The figures show force as a function of time from the entry to the end of the push phase.
Image 1a. The average propulsive (green), lateral (yellow), and vertical (red) forces produced during a stroke—long-distance swimmer. Image 1b. The average propulsive (green), lateral (yellow), and vertical (red) forces produced during a stroke—sprinter.
The stroke of a typical distance swimmer is characterized by a fairly long glide phase, during which all forces are minimal (Image 1a). Force production is maximized at the end of the stroke, and almost all of it should be propulsive (green). Sprinters, on the other hand, need to produce propulsive force right at the beginning of the stroke. Both the propulsive force and the stroke rate should be maximized even at the expense of optimal direction. It’s very common to see sprinters produce a large vertical force (red) at the beginning of their stroke. Even a substantial drop in the force at the end of the stroke is not critical, because the strokes overlap heavily thanks to a high stroke rate.
With SmartPaddle, the feel for the water becomes visible both to the swimmer and the coach. You can identify possible problems in stroke technique and verify the effect of corrective actions. For example, the force graph of a developing swimmer is presented in Image 2. It doesn’t do the swimmer much good if water is pushed in wrong direction (in this case lateral), especially during the strongest part of the stroke. Losing the force at the end of the stroke is also very common with developing swimmers.
Image 2. The average propulsive (green), lateral (yellow), and vertical (red) forces produced during a stroke—developing swimmer.
What Is the Difference Between a Distance Swimmer and a Sprinter?
As could be seen in Images 1a and 1b, the force graphs for a distance swimmer and a sprinter are quite different. Aside from the athletes’ pure physiological capabilities, a major difference is that swimmers usually master either the sprinting technique or a technique more suitable for longer distances. When swimmers try to apply their technique in the wrong distance, their performance is invariably below their true potential.
In a pool, my daughter was considered a pure sprinter excelling in 50m butterfly—but she couldn’t even finish a 200m distance despite several trials. In track and field, however, she was an adept long-distance runner while her sprinting performance was clearly below average. It was obvious to us that she needed to develop different swimming techniques for each distance to compete in them successfully.
According to our measurements, her greatest asset in sprinting distance was her powerful dolphin kicks. To use them even more efficiently, she needed to increase her stroke rate. She could achieve this by decreasing the force she applied with her hand strokes.
Using a high stroke rate and a powerful kick was not an option, however, if she wanted to finish a 200m race. Unfortunately, decreasing the stroke rate created a problem with a huge drag. But she could overcome this problem by adding a third kick in her stroke. During the third kick, she focused on attaining an extremely streamlined body posture, which made starting the hand stroke very easy. Using this technique not only made her strokes lighter but also decreased the number of strokes by 40%. After swimming the first 150m with the 3-kick technique, she could switch to a 2-kick technique for a very powerful last lap.
Which Is More Important—Pulling or Kicking?
This is really a trick question, as the answer depends on the swimming stroke, the distance, and the individual swimmer. Moreover, to be very successful, the swimmer has to master both. What’s often overlooked, though, is that the kicks can make the pulling much easier. Or haven’t you noticed that, when using fins, your feel for the water improves substantially?
The effect of the kicks is most pronounced in the catch phase when the swimmer’s velocity is at its lowest. Many swimmers pull their hand too fast at the beginning of the stroke and don’t gain much traction. With kicks, the swimmer’s velocity increases during the catch phase, and it’s much easier to get a good hold on the water. Consequently, the effect of the kicks is most important for swimmers who otherwise struggle with their catch phase.
With SmartPaddle, you can easily measure improvement in the hand strokes by conducting the test with and without kicking. According to our measurements, swimmers can produce approximately the same amount of propulsion at a certain stroke rate, regardless of whether they’re kicking (Image 3a). The difference is that with kicking, the strokes are significantly longer, and thus the speed is substantially higher (Image 3b). The effect of the kicks is pronounced at a low stroke rate because the speed variation is also greater.
Image 3a. The average propulsive force produced with and without kicking presented as a function of the stroke rate. Image 3b. The average stroke length with and without kicking presented as a function of the stroke rate.
What Happens After a Few Hundred Meters?
Have you ever wondered why you seem to hit a wall in long-distance swimming? In the beginning, swimming seems easy, and you make good time without much effort. After a few hundred meters, however, your arms seem to go numb, and each stroke is a struggle—even if you’re not particularly winded. Technique drills and video analysis don’t offer much help since these are always done when you’re fresh. That’s not when you experience the problems.
SmartPaddle provides a view of the changes taking place during long-distance swimming. For example, the impulse of each individual right-hand stroke is presented in Image 4 during a 14-lap test. In this case, the propulsive impulse (green bar) in lap 6 (between 40-50 strokes) decreased by half from the original. After that, it remained more or less constant for the rest of the test. Because the right hand was much weaker than the left, the speed of the swimmer always decreased when he pulled with it. As the right hand got weaker, the speed variation increased substantially, and the swimming became even more inefficient.
Image 4. The development of the right-hand impulse in a long-distance freestyle.
The swimmer was instructed to focus only on maintaining the force in his right hand. He also balanced his swimming by adding kicks to help the catch phase of the right-hand stroke. Within a week, the impulse of the right hand increased by 63%.
The swimmer improved his time in a 400m trial by more than 20 seconds while maintaining exactly the same stroke rate. This development was possible once the swimmer became aware of the weakness in his technique.
How it Works
SmartPaddle is a wearable sensor that measures the force, speed, and orientation of the swimming strokes. SmartPaddle shows whether the swimmer can apply the right amount of force in the right direction at the right time.
The SmartPaddle sensor is attached to the swimmer’s fingers, and the measurement is carried out with a mobile app (Image 5). You use the app in the same way as a regular stopwatch—it recognizes strokes and laps automatically and calculates a detailed analysis of the technique. The feedback is available immediately at the pool deck from the web service. All data is stored in a database for review, comparison, and follow up.
Image 5. Carrying out the measurements with SmartPaddle.
Time-Dependent Force
Force graphs presented in Images 1 and 2 give a good overview of the swimmer’s feel for the water. With time-dependent force data, the swimmer can visualize, for example, how the propulsive force varies during the measurement. As I demonstrated previously, variation may be an important source of inefficiency.
In Image 6, you can see that in this test, the right hand produces significantly more propulsion than the left. When the propulsive force is above the average force, the speed of the swimmer increases. Conversely, when propulsive force is below average force, the swimmer decelerates. Therefore, the speed of the swimmer increases during the right-hand stroke and decreases during the left. Thus, the average force provides information on the speed variation, which is one of the major reasons for losses in swimming performance.
Image 6. Time-depended propulsive force compared with the average force in freestyle.
The possibility of downloading the time-dependent force data also lets the user combine the force with advanced video analysis tools (Image 7). By creating a force overlay into a video, it’s easy to visualize and observe the propulsive, lateral, and vertical force components. Such a view provides unprecedented clarity to the force measurements.
Image 7. Time-depended propulsive forces overlay with underwater video.
Conclusions
As a scientist and an engineer, I value empirical testing over beliefs. Coaching still requires us to make educated guesses, but I prefer to base them on hard data. I find the motivation to train increases when the results are clearly visible. The better the swimmer is aware of their strengths and weaknesses, the easier it is to prepare for competitions—and being well prepared is the best source of self-confidence needed for winning.
When I first started testing the SmartPaddle, my daughter’s best swimming stroke was 100m butterfly, where she was ranked 21st in her age group in Finland. In two years, she became the Finnish junior champion in 50m butterfly, reached 2nd position in 100m and 3rd in 200m. She also took several podium positions in medley as well as backstroke. She might have reached that level without testing, but I never trust luck when there is another option.
5 min read
4 min read
6 min read
2 min read
3 min read
10 min read
4 min read
5 min read
3 min read
3 min read
2 min read
20 min read
The conference has issued CEUs every year and will be working with several associations to have them credential in March 2021. Our conference program usually needs to be fully designed with all speakers in place for us to be able to apply for the credentials. 
