On June 22, 2006, we tracked two 4-on-4 basketball games. These videos are from the first game that day. Each shows subsequent stages in processing, from (almost) raw data to full generated tracks. Note that one of the lasers was knocked over at around 18 minutes into the game. It is clear to see the effects of this as the laser was reset. Also, for the rest of game, the laser is noticably not aligned, yet the tracker still does well.
 (click the picture for a clip) |
In this video, the raw data has been spatially and temporally registered. The dots are color coded based on which laser generated the hit, while the large circles represent the lasers. The large room contains two courts; our game was on the left court. | |
 (click the picture for a clip or here for the full game) |
The data from the previous video was run through our background subtractor to remove all data not related to the 8 players. This is the data that is going to go to the tracker for tracking. | |
 (click the picture for a clip or here for the full game) |
This video shows the tracks that were generated by our tracker from the above data. | |
 (click the picture for a clip or here for the full game) |
In this video you can see the original background subtracted data (small dots) with the tracks (larger open circles) overlayed, to show the effectiveness of the tracker. |