ISU Congress 2022 (6-10 June)

[Perelandra]

@Sun_Rise

Yes, correct ISU congress discussion about the feasibility study for using the Sportlogiq tool was very wooly. ISU could decide that the use of technology is not feasible because it may an embarrassment to them when it exposes huge discrepancy between how skating programme has been judged objectively using the tech and the subjective scoring of the judges and their biases

There was no discussion about how or when the study would be implemented. Any report of methodology, results, conclusion would not surface until the next congress and perhaps might even be buried if the study demonstrates poor judging practice

 

That is why it is important that other forms of tech and expert analyses to be published on social media platforms. Allow the results to be viewed and let the interested public come to their own conclusions. 

ISU need to be aware that their practices cannot stay hidden and that people are accumulating knowledge about judging and scoring and becoming well informed and that overwhelming public opinion will matter to them when people make their own views count. If ISU become aware that they are being scrutinised and that their reputation is at stake and that their commercial interests hinge on the credibility of their public image they may take steps to reduce poor practice

As science teaches us, the act of observation can alter the result.

 

The choice of Sportlogiq. The following is conjecture only. The tool for the study is a Canadian brand. ISU must be aware of Yuzu's thesis. They may have illogically chosen a non Japanese brand because they imagine that it will not have any input from Yuzu's study. However, although Yuzu used equipment from Waseda which may well have been Japanese, it is the algorithm for analysis which is the keystone for the thesis. As far as anyone knows Yuzu has not launched a commercial product, nobody except Yuzu +/- team know whether he has been approached about using his algorithm for a larger trial/study. It would be ironic if the analytical algorithm used by Sportlogiq is base on the one created by Yuzu

 

Hope you don't mind posting on this thread, you might want to discuss here because this is related to ISU congress and not rant. The salt thread is archived regularly and this is a subject which may warrant further discussion as more information appears

[Anabel]

8 hours ago, Perelandra said:

@Sun_Rise

Yes, correct ISU congress discussion about the feasibility study for using the Sportlogiq tool was very wooly. ISU could decide that the use of technology is not feasible because it may an embarrassment to them when it exposes huge discrepancy between how skating programme has been judged objectively using the tech and the subjective scoring of the judges and their biases

There was no discussion about how or when the study would be implemented. Any report of methodology, results, conclusion would not surface until the next congress and perhaps might even be buried if the study demonstrates poor judging practice

 

That is why it is important that other forms of tech and expert analyses to be published on social media platforms. Allow the results to be viewed and let the interested public come to their own conclusions. 

ISU need to be aware that their practices cannot stay hidden and that people are accumulating knowledge about judging and scoring and becoming well informed and that overwhelming public opinion will matter to them when people make their own views count. If ISU become aware that they are being scrutinised and that their reputation is at stake and that their commercial interests hinge on the credibility of their public image they may take steps to reduce poor practice

As science teaches us, the act of observation can alter the result.

 

The choice of Sportlogiq. The following is conjecture only. The tool for the study is a Canadian brand. ISU must be aware of Yuzu's thesis. They may have illogically chosen a non Japanese brand because they imagine that it will not have any input from Yuzu's study. However, although Yuzu used equipment from Waseda which may well have been Japanese, it is the algorithm for analysis which is the keystone for the thesis. As far as anyone knows Yuzu has not launched a commercial product, nobody except Yuzu +/- team know whether he has been approached about using his algorithm for a larger trial/study. It would be ironic if the analytical algorithm used by Sportlogiq is base on the one created by Yuzu

 

Hope you don't mind posting on this thread, you might want to discuss here because this is related to ISU congress and not rant. The salt thread is archived regularly and this is a subject which may warrant further discussion as more information appears

For me, that is the key. Being the center of attention would make them feel compelled to do something and not say that they are occasionally studying it.

The technology is very, very advanced. The televisions use some for their broadcasts, so the only thing that the ISU lacks is the desire to work on it.

[Perelandra]

There are many expert and motivated Satellites who have already accessed some of the technological tools with which to analyse skating footage.

It will be really fascinating to see the motion capture tool used with scores derived from the tech vs scores awarded by judging panels.

It will equally interesting to see what kind of reactions these comparison scoring tool draw from people on social media platforms

 

Small fan folk don't need to plot against their nefarious ways, ISU are masterminding their own downfall, with their corrupt behaviour and poor practice

[Anabel]

And that's the funniest thing, it's not that some fans wish that things goes as they want, it's the need for the ISU to abandon bad practices and restore dignity to the FS. 

[Perelandra]

The sporting world has viewed the dreadfulness in FIG camp. They had to clean up and restore their reputation. It may be that ISU takes heed of the letter campaign and motion scoring. 

Being aware that there is scrutiny and judgement may encourage a clean up act. If not then boycott events unless Yuzu is attending. If nothing else the accountants will soon explain the truth about income stream.  If ticket sales and paid TV subscriptions plummet without Yuzu vs scramble for tickets and tv subs rocket with Yuzu, then the revenue from broadcasting rights, advertising and sponsorship follows, the message will get through. Equally, tarnished reputations quickly come to the attention of sponsoring parties, the will want distance between themselves and organisations which are negatively viewed. Once organisations feel financial pressure, the need for cash flow often wakes them up.  It's a long game but if it yields the desired result, better and accountable behaviour and practice. Then it will have been a worthwhile campaign. If they can't improve then everyone will see this and make their own decisions about continued complicity supporting a corrupt organisation or leaving them to rot. 

[Sun_Rise]

22 hours ago, Perelandra said:

They may have illogically chosen a non Japanese brand because they imagine that it will not have any input from Yuzu's study. However, although Yuzu used equipment from Waseda which may well have been Japanese, it is the algorithm for analysis which is the keystone for the thesis. As far as anyone knows Yuzu has not launched a commercial product, nobody except Yuzu +/- team know whether he has been approached about using his algorithm for a larger trial/study. It would be ironic if the analytical algorithm used by Sportlogiq is base on the one created by Yuzu

 

Yuzuru used Perception Neuron 2 from Noiton, a Chinese company (they have their offices also in Tokyo, Miami, LA). This is a system with really small, non-invasive sensors, currently offered as Perception Neuron 3 branded as the world's smallest motion capture system. 

 

However, altogether, the way he applied the system may be unique. We still don't have a full version of his thesis and maybe there is a very good reason for it. 

[Perelandra]

Thanks for clarifying. It was not clear, whether the equipment used for the study had been borrowed from Waseda equipment library or Yuzu had purchased himself.

 

Either way it is the algorithm/ software for analyses and methodology of the study which is the important aspect for developing the AI for use in competition.

Quite right, it is also the intellectual property of Yuzu, hence, only partly published material. How he chooses to develop and apply is his business, alongside his collaboration team.

As stated previously, I*U are being deliberately obtuse about the use of technology. The mere fact that even something quite ubiquitous as slow motion nor multi camera angles are not used is nonsense. This is so well known amongst competitors and coaches that it has been mentioned in skating circles that skaters often perform elements in judges 'blind spots' to avoid being called out on less strong aspects of their performance.  This has clearly been exploited for ages because judging is purely by eye and can miss so many details. 

Hopefully the realisation that the wider public have noticed and are watching this intently will lead to change of practice. Alternatively,  they won't care and essentially flush their own remaining shreds of reputation and credibility down the loo and be gone, swept aside for something better as a sports governing body.

 

Sad to say, but I hope that Yuzu will never be sucked into this corrupt organisation in any manner

If he coaches, let it be for consultation like technique masterclasses or sports mental game type of work. Otherwise if this dreadfully managed and run organisation somehow avoids extinction they would try to punish Yuzu's students. He would not forgive himself if he thought he'd brought ill treatment upon another person because he has such empathy.

[Sun_Rise]

Fujitsu comes as another obvious candidate and they offer yet another motion capture technology currently implemented to gymnastics, which is a 3D laser. Reportedly, the ISU has also been at least reviewing the technology to possibly adapt it to fs (but, again, it's all bits and pieces we're receiving).  

 

With gymnastics it was clear that it was about Japanese brand first pairing with Japanese feds, then with the International Gymnastics Assoc. before the OG in Tokyo. Japanese government supported it to present the country as a leader in introducing innovative technology to sport. 

[Anabel]

1 hour ago, Sun_Rise said:

 

Yuzuru used Perception Neuron 2 from Noiton, a Chinese company (they have their offices also in Tokyo, Miami, LA). This is a system with really small, non-invasive sensors, currently offered as Perception Neuron 3 branded as the world's smallest motion capture system. 

 

However, altogether, the way he applied the system may be unique. We still don't have a full version of his thesis and maybe there is a very good reason for it. 

I was just thinking that. Many movies are made today using that kind of technology. In the case of movies, the number of sensors is large because they need to capture movements in more detail.

 

I was also thinking about the IceScope you named. If it measures the speed at which the skater rotates, it should also be able to tell when the skater starts the rotation due to the speed change, as well as when the skater land it. I'm not sure if it could also evaluate the posture. It's all guesswork on my part.

 

It would be ironic if the ISU is "forced" to use AI, they would turn to Yuzu and his studio to apply it. I don't know if Yuzu would like to turn his thesis into his professional work when he retires.

[Wintek]

4 minutes ago, Anabel said:

I was just thinking that. Many movies are made today using that kind of technology. In the case of movies, the number of sensors is large because they need to capture movements in more detail.

 

I was also thinking about the IceScope you named. If it measures the speed at which the skater rotates, it should also be able to tell when the skater starts the rotation due to the speed change, as well as when the skater land it. I'm not sure if it could also evaluate the posture. It's all guesswork on my part.

 

It would be ironic if the ISU is "forced" to use AI, they would turn to Yuzu and his studio to apply it. I don't know if Yuzu would like to turn his thesis into his professional work when he retires.

The motion caption device tested by Yuzu for his thesis is that kind of sensors used for movies, but smaller (12.5mm x 13.1mm x 4.3 mm, ~1g). He used them to capture the rotation of his jumps and it seems it worked OK overall. Perhaps some sensors, not all of them could be used for critical aspects that cannot be clearly measured/captured by cameras (even by IceScope type of tech). For example, a cool finding is that the sensors can show the contact point of the foot, which could help, for example, determine which side of the foot (i.e. edge) was used for take off.

[Sun_Rise]

26 minutes ago, Anabel said:

I was also thinking about the IceScope you named. If it measures the speed at which the skater rotates, it should also be able to tell when the skater starts the rotation due to the speed change, as well as when the skater land it. I'm not sure if it could also evaluate the posture. It's all guesswork on my part.

WIth rotations it is already possible with high quality slow motion, but even this is not utilised by the judges. It is also possible to have a 3D picture based on material coming from videos. In Saitama there were two 4k cameras. 

 

@Wintek super cool finding and yes, inertia sensors have their advantage over videos and could possibly detect prerotation or other aspects of the take-off (as well as the whole jump/movement). 

 

Btw imagine Yuzu records all the fs movements with these sensors and creates a database of Yuzuru Hanyu skating!

 

EDIT: Now I thought he could have done that already, "knowing" him. 

[Perelandra]

The Fujitsu laser tech was considered but the sporting arena for gymnastics is much more circumscribed and predictable compared with the competition rinks. Therefore it's use for skating comps was limited.

There is an impression that as this tech had been successfully used in gymnastics, ISU will turn away from this. They seem quite determined to hide from this issue for as long as possible. They are surely aware that their corrupt practices will soon be called out