Hawk Eye Challenge System has Arrived!

Discussion in 'Rules / Tournament Regulation / Officiating' started by Tactim, Apr 4, 2014.

  1. craigandy

    craigandy Regular Member

    Joined:
    Aug 9, 2011
    Messages:
    2,363
    Likes Received:
    18
    Location:
    UK
    They will just take the centre of the cork no? The cork is all they have to concerned about with regards to contact point
     
  2. jyeung

    jyeung Regular Member

    Joined:
    Mar 19, 2008
    Messages:
    250
    Likes Received:
    124
    Location:
    VANCOUVER
    I think the main reason for implementing the Hawk-eye system is to take line judge partiality out of the equation. As long as the system perform at a relatively consistent level, tolerance of 1mm or 5mm is not as important, since both sides would be subjected to the same standard! Whether it is a good value for the cost is another matter!
     
  3. craigandy

    craigandy Regular Member

    Joined:
    Aug 9, 2011
    Messages:
    2,363
    Likes Received:
    18
    Location:
    UK
    Well that's one way to look at it, another way - If it has tolerance of 5mm then it is not really solving the problem that they want to solve, which is when a player decides to question a possible bad call BWF wants an actual truthful answer or what's the point?
    What Hawkeye does offer is a is a very expensive but great side show but not a solution, unless the intent is just to make the sport more exciting for fans.
     
  4. RedShuttle

    RedShuttle Regular Member

    Joined:
    Oct 18, 2008
    Messages:
    1,688
    Likes Received:
    242
    Location:
    Western Hemisphere
    No. It's the whole shuttle. Bottom or side of the cork, or the feathers.

    Depending on the position of the shuttlecock in flight, any part may touch the floor first.
     
  5. RedShuttle

    RedShuttle Regular Member

    Joined:
    Oct 18, 2008
    Messages:
    1,688
    Likes Received:
    242
    Location:
    Western Hemisphere
    From the BWF news release, they are more interested in leveraging the Hawk-Eye brand appeal.

    They seem to think the spending in Hawk-Eye will translate to fan appeal and bring people to badminton.
     
  6. craigandy

    craigandy Regular Member

    Joined:
    Aug 9, 2011
    Messages:
    2,363
    Likes Received:
    18
    Location:
    UK
    No I'm pretty sure it's not possible for the initial contact point to be anything but the rounded area on the cork. Unless It is heavily tumbled but even then it's never happened. If you do the maths you will find out it is not possible. The space is too small for it to be possible if the drag difference were the same(maybe just diagonally) but the difference in drag between cork and feather is big too so no chance.
     
  7. captaincook

    captaincook Regular Member

    Joined:
    Mar 23, 2012
    Messages:
    272
    Likes Received:
    0
    Occupation:
    captain
    Location:
    New York
    What were the competing systems ? (aside from Hawk Eye)
     
  8. pcll99

    pcll99 Regular Member

    Joined:
    Nov 19, 2009
    Messages:
    8,601
    Likes Received:
    589
    Occupation:
    Cylon
    Location:
    N/A
    am i right to say that hawkeye's result is the statistical probability of the likely landing site based on the observed trajectory?

    hawkeye's result is not an observation but a calculation?
     
  9. Oldhand

    Oldhand Moderator

    Joined:
    Nov 21, 2005
    Messages:
    6,842
    Likes Received:
    107
    Occupation:
    Broadcast Systems Integration
    Location:
    Asia
    You are spot on.

    It's a calculated landing point.
    It might or might not be precisely the same as the actual landing point.
     
  10. pcll99

    pcll99 Regular Member

    Joined:
    Nov 19, 2009
    Messages:
    8,601
    Likes Received:
    589
    Occupation:
    Cylon
    Location:
    N/A
    If I am right, then it's not worth $60,000 per tournament to have hawkeye.

    Slow motion replay (at above 1000 fps) is sufficient.

    Having more cameras from different angles and higher capture frame rate would be better investment.
     
  11. Oldhand

    Oldhand Moderator

    Joined:
    Nov 21, 2005
    Messages:
    6,842
    Likes Received:
    107
    Occupation:
    Broadcast Systems Integration
    Location:
    Asia
    Nope.
    It would be costlier if you want broadcast quality.

    However, I agree with you that our cash-strapped sport doesn't need such a costly system right now.

    Hawk-Eye is much more than an electronic line-judge or dispute adjudicator.
    The system has several add-ons that can add punch, flavour and colour to the broadcasts.
    For now, it isn't clear if the BWF intends to make use of those 'additional' features.

    Also, a camera that shoots 1000 (or more) frames per second could be very cheap or very costly.
    One of those Casio EX series cameras would probably cost a few hundred dollars.
    But a professional unit like the Phantom Flex would cost in the region of $100,000.
    (The costliest model - the full Flex4K - costs over $170,000.)
    The difference in price is actually the cost of the (vast) difference in quality.

    The Hawk-Eye system relies on not just the shuttle's position-over-time as seen by a 2D camera (in fact, it computes the shuttle as a stable flight-point over which the shuttle's in-flight profile is overlaid later for television replays) but on the shuttle's continuously changing position-over-time-and-space in a mapped 3D environment.

    To put it in another way, the system relies on the tracking devices placed around the court to pinpoint the shuttle's position at any given point in time and then constantly compares the changing values against an unchanging reference (the unchanging reference = the 40mm lines on the badminton court) to compute the path in 4D.

    In the simplest terms, Hawk-Eye tracks in 2D (x,y), maps in 3D (x,y,z), then predicts in 4D (x,y,z + time) :D
     
  12. visor

    visor Regular Member

    Joined:
    Dec 7, 2009
    Messages:
    15,956
    Likes Received:
    1,567
    Location:
    Vancouver, BC
    Very well put.

    Now if only I can program Hawk Eye to predict stocks movement, I'll be laughin'... :D
     
  13. pcll99

    pcll99 Regular Member

    Joined:
    Nov 19, 2009
    Messages:
    8,601
    Likes Received:
    589
    Occupation:
    Cylon
    Location:
    N/A
    Two observations:

    (1) the position of the courts, nets and lines are relocated from quarter finals, semi finals to finals. The hawkeye system has to be recalibrated every night, right?

    (2) the number of people inside the arena/stadium affect the trajectory path of the shuttle by quite a bit, doesn't it? At 2:30pm, there could be 2000 people, but by 8pm, there could be 4000. Even assuming the air-con is constant throughout the day, doesn't the hawkeye need to recalibrate very often?
     
  14. craigandy

    craigandy Regular Member

    Joined:
    Aug 9, 2011
    Messages:
    2,363
    Likes Received:
    18
    Location:
    UK
    Point 1 is interesting that does seem like it will be a struggle. another thing to tag on to this are the courts layed with enough precision from front to back side to side?

    (2) I Think as oldhand said it tracks in 2d(to an extent) so it has a good bit of information to go on, whether the hall heats up, just like if the shuttle is struck faster it coordinates information from the cameras. So basically unless in the last 0.003seconds(just a guess) of its flight the heat rapidly changes or there is a big gust wind in that small time frame, then the system should work as it always does.
     
  15. Cheung

    Cheung Moderator

    Joined:
    Apr 25, 2002
    Messages:
    21,081
    Likes Received:
    2,867
    Occupation:
    wannabe badminton phototaker
    Location:
    Outside the box
    Given these slight drawbacks to the system, and the high cost relative to prize money, it's going to be hard to justify hawkeye in on a regular basis.

    All the umpires and linesman work on a voluntary basis.
     
  16. craigandy

    craigandy Regular Member

    Joined:
    Aug 9, 2011
    Messages:
    2,363
    Likes Received:
    18
    Location:
    UK
    I am pretty much convinced that the justification for Hawkeye is to make badminton look flashy and appear professional and for it's added excitement for fans (which may all bring in audiences). Not so much for the actual line call decision benefit.
     
  17. pcll99

    pcll99 Regular Member

    Joined:
    Nov 19, 2009
    Messages:
    8,601
    Likes Received:
    589
    Occupation:
    Cylon
    Location:
    N/A
    thanks. good observation.

    What I meant was this. Increase in the number of spectators from 2000 to 4000 may affect the "calculation" or prediction of the hawkeye system, which was presumably pre-calibrated the night before (or early in the morning).

    But if the system can re-calibrate itself on the fly, that would be great!!!
     
  18. craigandy

    craigandy Regular Member

    Joined:
    Aug 9, 2011
    Messages:
    2,363
    Likes Received:
    18
    Location:
    UK
    Right correct me if I'm wrong because it would be nice to get this straightened out.

    Hawkeye explained in a really basic way-
    It has a bunch of cameras that don't have crazy high fps. Then what happens as the shuttle is flying across the court the cameras all talk back to the central "unit" which processes this at every frame. So camera 1 catches the shuttle at every frame so first frame shuttle is 1.5metres away at an angle of 10degrees etc etc. Camera 2,3,4,5,6.... are all doing the same thing. All of them together give the info from each frame making the system more powerful than just one camera at "x" fps and also see's it from more angles.

    Therefore whether 2000 then 4000 people arrive or the air con is switched on or off has no bearing or relevance.

    The calibration only comes into play on set up or if the cameras move or the court move, because if that happens each camera has to be set pointing and tilted at exactly the correct angles to correlate with system that interprets it. With multiple cameras all providing information each one has to be perfectly set for the calibrated distance and angle to make sense
     
    #58 craigandy, Apr 9, 2014
    Last edited: Apr 9, 2014
  19. RedShuttle

    RedShuttle Regular Member

    Joined:
    Oct 18, 2008
    Messages:
    1,688
    Likes Received:
    242
    Location:
    Western Hemisphere
    As long as the cameras collectively have good lines of sight (from various angles and heights) and are set firmly (no vibrations from wind, fan activities, etc), their exact locations shouldn't matter.

    The calibration is for the mathematical model to correctly interpret the inputs from the cameras, probably by some sort of triangulation calculation. Once the cameras are in place, they can do that with test targets at strategic locations.
     
  20. craigandy

    craigandy Regular Member

    Joined:
    Aug 9, 2011
    Messages:
    2,363
    Likes Received:
    18
    Location:
    UK
    Yeah that's what I was trying to say:D, I was just trying to state they cant use the same calibration with the same cameras positions from QF to F.
     

Share This Page