i am sorry. i missed "observe high speed video footage link". may i ask u which post of mine u are talking about?
Not sure about cricket, but for tennis, have you noticed how big the ball mark is? Unless they change the rule of badminton which goes by the first point of contact, the equivalent "ball mark" is tiny compared to that in tennis. For example, if the shuttlecock is dropping almost vertically (e.g. a high clear to the back court), even a part of shuttle cork (say 1/4) is over the line, the tip of cork will land first outside the line. That's an "out". But a similar situation in tennis would be an "in" on the count that the whole cork will collapse over the line and part of it will catch the line.
It's easy to get carried away with this fps issue, IMO. Even a 180 or 240fps data is sufficient to track and project the path of the object, IF the algorithm is correct. If Hawkeye is doing 1000 fps that is super! The problem is that there are too many variables and the tracking resolution needs to be accurate. Most filters that high-end broadcast/film equipment use in post-processing (or reviewing) are fully capable of magnifying and sharpening the outline/image to ascertain any type of deviation from the mean. This will tell the system if there is any spin or drift at play. How the system differentiates between spin and drift (and this can be a crucial factor in some cases) will allow it to factor both into the equation. As for the point of impact being the cork (or tip of the cork as RedShuttle mentioned above) even this is a relatively simple issue. Camera triangulation allows the system to ascertain the line drawn through the centre of the shuttlecock and the point of impact can be allocated an active area with a radius of say, 2mm or 4mm or whatever. In tennis moreover, the ball tends to squash due to the force of the ground acting on it, and that is why the system needs to make allowance for this and accordingly enlarge the active area of impact.
It was post #34, you provided a link click it and if you read the article that sentence is highlighted, here it is though https://vimeo.com/40523215
That is supposed to be how it works, but maybe you are right for badminton?? The idea is relatively straightforward. At Wimbledon this year, Hawk-Eye will use a system of 10 cameras to photograph the ball in flight. A computer triangulates the ball's position in three-dimensional space and repeats the process for the next frame. ”It then joins the dots,” says Hawkins, who talked to IEEE Spectrum after a day on Centre Court at Wimbledon setting up the system for this year's tournament. The computer then produces an animated picture of the ball's trajectory in a virtual tennis court, showing how the ball moved in the air and where it bounced. http://spectrum.ieee.org/computing/software/hawkeye-in-the-crosshairs-at-wimbledon-again
I don't think that is right, The article certainly was not clear but I think that was a camera review that the ITF took to proof Hawkeye was getting the calls right. This is my suspicion anyway because nowhere have I seen a definitive statement on the fps of their cameras. There is this blog http://videosportsanalysis.blogspot.co.uk/2009/09/how-accurate-is-hawk-eye-system-for.html It claims 60fps per camera and when put together 500fps He also expresses the same concern we are having about badminton speed The camera capture rate of 500 frames per second may be too slow for the hard hitting tennis game of today. Andy Roddick has recorded a fastest serve at 155 mph and ground strokes are often hit with a similar speed. Assuming a capture speed of 500 frames per second or 1 frame every 0.002 seconds, the ball will have moved about 139 mm for every video frame. This may not be sufficient data to determine the position of the ball to within 1 mm. For sure re: animation
As many of you would know, there is a more discussion and good information to be found on another, earlier thread on this subject (DRS) http://www.badmintoncentral.com/for...play-system-to-check-disputes-over-line-calls
I think Hawk-eye was wrong on the sideline call in the XD match. The touch down point was outside the line. The animation showed a giant ball mark that caught the line. The cork on the shuttlecock does not compress that much to make such a ball mark.
Check this out: http://www.youtube.com/watch?list=P...eature=player_detailpage&v=RbIWfM7XkJ0#t=3401 Hawkeye's reconstruction for the challenge called by Simon Santoso. If you rewind to 10 seconds back you will also see the video replay of the smash by SS. I think the shot was in and well on the line. Hawkeye's animation does not show any dip that we would be accustomed to seeing as the shuttlecock loses its momentum, and the result shows the bird just about clipping the outer edge of the line. That is a worrying sign.
Your link doesn't start at the right spot. But I think I know which challenge you're referring to as I just watched it. Imho I thought Hawk-Eye got it right. I thought it was out.
Seriously how is this ever better than a highspeed camera's slow-mo replay? The true picture is hidden and the simpletons are brushed off with an antiquated 3D computer animation. Have never seen a cork (on a match shuttle) that's bigger than the line either
Please be reminded that the Hawk-eye animation is not the reality but just an illustration of a scenario conjured up by the system. The animation is not a proof that Hawk-eye is correct but is just what Hawk-eye perceives. Given the peculiarities of the shuttlecock flight characteristics, I am not convinced that the Hawk-eye system recreates the reality faithfully. They need to show some footages of super slow-mo video super-imposed with Hawk-eye animation to convince me. By the way, it seems strange that the middle service line is not covered by Hawk-eye. Once set up, what so hard about covering all lines on court?
All the service lines don't have to be covered because serve shots are much slower speeds, so they're easier for the line judges to see.
There was one challenge that wasn't too accurate yesterday in SS vs LCW. In was actually the second point in the match in which SS played a slicing straight drop to the right side line which was challenged by SS for being called out. From video replay the shuttle I think definitely hit the line right on the spot, but hawk eye showed that the shuttle hit INSIDE the line, which was not the case. In the end, hawk eye made the right call though not by the margins it was showing on the animation. So, verdict: Hawk Eye needs more work and calibration and testing. I don't think it's ready for matches, YET. They also need to modify it so that the space which the shuttle hits is more circular rather than a squashed oval shape because it uses there same graphic as if it was a tennis ball hitting the ground and being compressed as it hits the ground.
is it just me or is it REALLY off, the two callenges ive seen with LCW + KIDA and LCA + SANTOSO, the hawkeye actually put it on the WRONG side of the line lol...........
Yah, those graphics need to be accurate. Even if Hawk-Eye has an error tolerance of +/- 3mm, it should at least look like the slow mo replay if it wants to instill any confidence in it's decisions. Fortunately for it, the spectators don't know any better because they don't see the slow mo instant video replay overhead, only the Hawk-Eye computer graphic version. If i were BWF, I'd at least demand half my money back.
This is really well said. In tennis at least the antiquated 3d animation actually follows the path/angle that the ball took so is reasonably believable. In badminton they don't even attempt this. What they have done for badminton is just an insult. I mean it isn't solving the line call problem because we know there is tolerance in the system, and the entertainment value from the animation is terrible. They may as well pay a fraction of the money per court and get 4 circus clowns to come in, guess line calls and do routines in front of the crowd. They would achieve just as little as hawkeye.
