Title :
Bayesian Network-Based Customized Highlight Generation for Broadcast Soccer Videos
Author :
Kolekar, Maheshkumar H. ; Sengupta, Somnath
Author_Institution :
Dept. of Electr. Eng., Indian Inst. of Technol. Patna, Patna, India
Abstract :
Sports highlight generation techniques aim at condensing a full-length video to a significantly shortened version that still preserves the main interesting content of the original video. In this paper, we present the system for automatically generating the highlights from sports TV broadcasts. The proposed system detects exciting clips based on audio features and then classify the individual scenes within the clip into events such as replay, player, referee, spectator, and players gathering. A probabilistic Bayesian belief network based on observed events is used to assign semantic concept-labels to the exciting clips, such as goals, saves, yellow-cards, red-cards, and kicks in soccer video sequences. The labeled clips are selected according to their degree of importance to include in the highlights. We have successfully generated highlights from soccer video sequences.
Keywords :
Bayes methods; image sequences; probability; television broadcasting; video communication; audio feature; broadcast soccer video sequence; customized highlight generation; probabilistic Bayesian belief network; semantic concept-label; sports TV broadcasting; Bayes methods; Feature extraction; Games; Image color analysis; Semantics; Video sequences; Videos; Bayesian belief network (BBN); event detection; semantic annotation; soccer video indexing; sports highlights;
Journal_Title :
Broadcasting, IEEE Transactions on
DOI :
10.1109/TBC.2015.2424011