A fencing foil tip moves at roughly 62 mph. Human visual reaction time is 250 milliseconds. In epee, a double touch counts as a double touch only if both hits land within 40 ms of each other.
You are physically incapable of seeing who scored without a machine.
This is why fencing has been broadcast-hostile for its entire modern history. Five sports have appeared in every modern Olympics (fencing, athletics, cycling, swimming, gymnastics), and fencing is the only one almost no one watches outside the Games. The scoring moment happens below the threshold of human perception.
Rhizomatiks and Dentsu Lab Tokyo have been working on this since 2012. The first version used physical markers on the blade. Useless for real competition, because top fencers refuse to modify their weapons.
The markerless version is the actual engineering breakthrough. At 4K resolution, a sword tip is only a few pixels wide. The blade bends and flexes mid-strike, so the geometry is non-rigid. Traditional computer vision dies on both problems. They trained YOLO-variant deep learning models to detect the tip frame by frame and reconstruct the trajectory in AR, live, while the point is still being fought.
First real-match deployment was the 72nd All Japan Fencing Championships in 2019. It is now the core tech behind the new World Fencing League. Miles Chamley-Watson built the whole product around this tracking system and pulled in Nike, Airbnb, Mercedes, Richard Mille, and CashApp as backers. None of that happens if the scoring is invisible.
Tennis got Hawk-Eye. Baseball got pitch tracking. Golf got shot trace. Each of those turned a niche broadcast into a billion-dollar property by making the invisible visible.
Fencing was the last major sport waiting for its visualization layer.
Good @nytimes report on how genocidal language has been normalised in Israel
"Calls for Gaza to be “flattened,” “erased” or “destroyed” had been mentioned about 18,000 times since Oct. 7 in Hebrew posts on X...
The cumulative effect, experts say, has been to normalize public talk of “erasing” the people of Gaza, ethnic cleansing, and the nuclear annihilation of the territory..."
‘Erase Gaza’: War Unleashes Incendiary Rhetoric in Israel https://t.co/b44m2szu7K
My work on Urdu technology, @matnsaz, was featured in Time! And to be profiled with others doing such amazing work, such as @ZeeshanNasar@Rekhta, it’s an absolute honor.
https://t.co/YmjhVv33Tg
@Hazen_ai will be at @Intertraffic Amsterdam, March 29 - April 1. Visit us at our booth 05.419 for the latest in video analytics for traffic safety. #intertraffic
جان سے لیکن گزرے تو
آئے پھر اپنانے لوگ
جب کوئی حسرت نہ رہی
بیٹھے پیار جتانے لوگ
چادر لے کر لپکے پھر
بُن کر تانے بانے لوگ
جیتے جی پوشاک نہ دی
پہنچے ہاں کفنانے لوگ
بوجھ بٹاٹے ڈرتے تھے
پیش کریں اب شانے لوگ
قصہ پاک ہوا گرچہ
چلے ہیں گو دفنانے لوگ
#ڈاکٹر_عبدالقدیر_خان#محسن_پاکستان
After two years of writing + editing, the (proof-read) hard copy is finally here! Thank you to everyone who helped with this book.
I'll tweet about a few unique things about this book.
To get a PDF copy of this book:
https://t.co/imyXY47ePH
#datascience, #MachineLearning
Thrilled to back @Tazahtech with an another amazing rockstar 🕺 team. I knew within 20 minutes of meeting Abrar & Mohsin in our conversation that this is the team that can execute and scale. @GFC_SEA @ZaynCapitalVC
https://t.co/xhMSPC5VTJ
At @NVIDIAAI#GTC21, we'll be presenting on SS32442: Can AI Make Our Roads Safer? Starts April 12 at 10 am (PDT, UTC - 7). Ensure you don’t miss it by registering for free at https://t.co/qTMvPbbXyq
🇵🇰LAHORI TWITTER🇵🇰
I’ve developed an aversion to eating out at fancy restaurants in Lahore.
Dhabas, khokhas, thelay, road-side eateries offer abundant taste and plenty of Lahori love.
Here’s a list of some of my favourites, along with google pins. Feel free to share!