Tennis Rally 

#Source:

D. Y. Fu, M. F. Chen, F. Sala, S. M. Hooper, K. Fatahalian, and C. Ré. Fast and three-rious: Speeding up weak supervision with triplet methods. In ICML, pages 3280–3291, 2020.


#Labels:

0: negative (the game is not rally segment)

1: positive (the game is rally segment)



6 Labeling functions

LFs: this dataset uses an off-the-shelf pose detector to provide primitives for the weak supervision sources. The supervision sources are heuristics based on the number of people on court and their positions. Additional supervision sources use color histograms of the frames (i.e., how green the frame is, or whether there are enough white pixels for the court markings to be shown).