Heya! I’m a PhD Student at Cambridge University specialising in the application of Machine Learning in Music, and an Orchestral/jazz composer, arranger, pianist, trombonist, and Guinness World Record holding juggler!
Hey there! Thanks so much for the question. So an interesting effect of converting asynchronous Siteswaps to synchronous notation is that, depending on how you allocate the new beat positions, you may have 1s being “compressed” to “0x”s. In theory these are performable, but some prefer avoiding this Siteswap value, and instead simply catching the ball in the opposite hand, or as a multiplex etc.
How do you compare the difficulty of juggling while sat like this? I juggle like this a lot at home due to my low ceilings and feel that I am better when stood up. Maybe it is good training? I am really not sure!?
Hey there! Yeah I would say that in general it’s a lot harder to juggle while sitting down as it’s equivalent to “isolated endurance”, in the sense that you can’t move around to account for bad throws. This is even more so the case with 7+ balls, as you require a lot more height and so the potential width of the throws given bad releases increases. Also, with 7+, you need a a greater vertical distance while catching and releasing to accelerate it to the right speed, which is hard to do when sitting down! I imagine it’s probably good training for getting more consistent for sure!
Hey there! Yeah it took probably a year or so to get from 100 catches with 5 to 100 catches with 7, although it varies depending on how much you practice, and how much you keep it up, as it can become quite physically demanding!! Are you learning 7?
@@jamescozensjuggling6868 Yea I understand there are so many variables. Not up to learning 7 just yet...I am still trying to master 5...and I am fairly solid with 3 in one hand but haven't tried 6 either other than a flash for fun. I started juggling 4 months ago in my late 30's. Much fun😁 I love watching people who really know their skill👍
Interesting! At 0:12 there is a frame where two balls are close and the tracking didn't work. Is the reason the code sought something more like a ball and found an eight?
Hey there! Good spot! So basically there’s two stages to object tracking, the actual object detection, and then assignment. This is a demonstration of object detection, where the task is simply to extract the likely ball positions in the frame. If there is partial occlusion (like at that moment you pointed out), the algorithm sometimes treats the detection as a single object. This is not an issue as during the assignment phase, a kalman like filter algorithm predicts the position of the partially hidden object, so 9 objects are always accounted for. Let me know if you have any specific questions 🙂
@@jamescozensjuggling6868 what is the delay between the moment the ball is in the position N and the moment you can get the data of this position? With such thing skynet will never miss a shot ☠️☠️☠️
Heya! Sorry for taking so long to reply haha - I’ve just been finishing off the paper for it. So yes the new version can pick up every Siteswap value including 0, by taking a general “trend” (looking at the catch frequency to determine if there are “missed” throws)
It'd be interesting to measure how far off each ball is from its "ideal" position (which I guess would have to be computed from some sort of average). I'm imagining seven side-by-side pictures of a square with a colored vector in it (moving over time), indicating the difference between the actual position of the ball vs. its ideal position. (Or possibly you could draw the vectors all atop one another, but I worry that may look confusing.) Maybe disappear them when they're in hand. It would definitely be interesting to see how errors propagate and are eventually resolved; if they move from one ball to another, one might want to look at sums of these vectors as well.