Abstract:We address the problem of modeling and classifying American Football offense teams' plays in video, a challenging example of group activity analysis. Automatic play classification will allow coaches to infer patterns and tendencies of opponents more efficiently, resulting in better strategy planning in a game. We define a football play as a unique combination of player trajectories. We develop a framework that uses player trajectories as inputs to MedLDA, a supervised topic model. The joint maximization of bot… Show more
“…However, previous work has demonstrated that successful tracking of football players under accurate video registration is still notoriously difficult [8,6]. Recent work seeks to relax the mentioned requirements for feature extraction [1,16], however they rely on several crucial assumptions that do not reliably hold in our web-service setting. A key part of their approach is video registration and background subtraction.…”
Section: Related Workmentioning
confidence: 99%
“…After extracting relevant video features, existing methods typically employ probabilistic generative models for play-type recognition, including a Bayesian network [9], non-stationary Hidden Markov Model [15], topic model [16], and mixture of pictorial-structure model [7]. These models are typically used for each video in isolation.…”
This paper presents a vision system for recognizing the sequence of plays in amateur videos of American football games (e.g. offense, defense, kickoff, punt, etc
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.